PLASMA NORADRENALINE LEVELS
AND
CARDIOVASCULAR SYMPATHETIC ACTIVITY
A thesis presented for the degree of
DOCTOR OF MEDICINE
in the Faculty of Medicine of the
University of London
by
Anthony James Ivor Scriven
BSc (King's) MB BS (Westminster) MRCP
Work carried out at the
Royal Postgraduate Medical School, London W12
and the
Centre Hospitallier Universitaire Henri-Mondor, Paris
1 ABSTRACT
The validity of plasma noradrenaline as a biochemical marker of cardiovascular sympathetic activity was assessed.
The ability of peripheral venous plasma noradrenaline to reflect small generalised increases in sympathetic outflow was investigated using tyramine to release endogenous noradrenaline from sympathetic neurones. This showed that the relationship between noradrenaline release and plasma noradrenaline was linear, but the increase in plasma noradrenaline associated with a 10-15 mmHg pressure rise was small, and not easily detected.
There was considerable variabilty between individuals' noradrenaline clearance, resting blood pressure, and pressor responsiveness; but no association between these factors and plasma noradrenaline levels. This suggested that plasma noradrenaline could detect short-term changes in sympathetic activity within a group, but was unsuitable for comparisons between individuals.
Further studies showed that tyramine raised blood pressure by stimulating cardiac contractility. Therefore, to assess the significance of plasma noradrenaline during regional changes in noradrenaline release, the effects of tyramine were compared with cold stress, which causes peripheral sympathetic activation. The results showed that the increment of plasma noradrenaline associated with a given rise in blood pressure was not constant, but appeared to vary with different regional
2 contributions to the pressor response. It was concluded that the pattern of regional sympathetic activity cannot be assessed by peripheral plasma noradrenaline. Furthermore, even within an individual, the ability of plasma noradrenaline to compare different sympathetic responses is in doubt.
A new method was devised to assess cardiac sympathetic responses by estimation of intracardiac noradrenaline turnover.
This was measured at rest, and gave results which have been confirmed by others, using different methods. The technique of coupled pacing produced a marked increase in cardiac noradrenaline release; this was accompanied by haemodynamic changes consistent with increased cardiac sympathetic activity.
Paradoxically, propranolol also caused increased cardiac noradrenaline release for reasons which were not clear. It was concluded that this method may be useful for more detailed assessment of cardiac sympathetic responses.
3 CONTENTS
Abstract
Chapter 1 Introduction
Chapter 2 The structure and function of the sympathetic
nervous system
Chapter 3 The role of the sympathetic nervous system in
Chapter 4 The relationship between sympathetic nervous
activity and the plasma noradrenaline concentration
Chapter 5 General methodology
Chapter 6 The relationship between blood pressure and plasma
noradrenaline levels after release of endogenous
neuronal noradrenaline by tyramine
Chapter 7 The effects of propranolol on the pressor response
to tyramine
Chapter 8 The effects of tyramine on cardiac systolic time
intervals before and after propranolol
4 Chapter 9 Tyramine and cold stress
Chapter 10 A method for the estimation of intracardiac
noradrenaline kinetics
Chapter 11 Cardiac noradrenaline release at rest and during
coupled right ventricular pacing
Chapter 12 The effects of propranolol on cardiac noradrenaline
release
Chapter 13 Conclusions
Acknowledgements
Appendix Assay instruction sheet
Bibliography
5 CHAPTER 1
INTRODUCTION
6 1.1 INTRODUCTION
The original and widely-held view of the sympathetic nervous system was that of the "fight or flight" reaction initiated by a sudden and massive activation of the adrenomedullary system leading to an outpouring of adrenaline from the adrenal medulla. The effects - a pounding tachycardia, deep rapid respiration, widely dilated pupils, piloerection, sweating and pallor -are well known to many. This dramatic defence reaction seemed to have little to do with everyday regulation of the arterial blood pressure.
However, the physiological role of the sympathetic nervous is now better understood, and it is clear that it is probably the dominant force in the control of the circulation. 'The primary function is to maintain tissue perfusion within set limits. This is achieved by control of cardiac output and systemic resistance, and requires maintainance of a fairly constant level of arterial blood pressure in the face of frequent changes of posture, physical activity and metabolic activity.
Appreciation of the normal role of the sympathetic nervous system in blood pressure control led to interest in what its
role might be in essential hypertension. Certain patients, usually in the earlier stages of hypertension, may have haemodynamic and sometimes somatic features suggestive of
7 excessive sympathetic nervous activity; thus, a 'neurogenic' hypothesis of essential hypertension held that excessive activity of the sympathetic nervous system might have a primary role in the initiation or maintenance of elevated blood pressure in essential hypertension.
In the early 1970s, the concept of neurogenic essential hypertension driven by a primary increase in sympathetic activity received considerable support from several reports that hypertensive patients had elevated levels of plasma noradrenaline compared with normotensive controls. These observations, which were highly controversial, stimulated an enormous research effort devoted to the measurement and significance of plasma noradrenaline in populations of hypertensive and normotensive subjects.
The central assumption common to all of these studies was that plasma noradrenaline was a valid biochemical marker of cardiovascular sympathetic activity. This assumption was based on very incomplete data. A small number of isolated organ studies had shown that the release of sympathetic transmitter was proportional to the intensity of nerve stimulation.
Subsequent work showed that a variety of physiological and psychological stimuli known to be associated with increased sympathetic nervous activity consistently produced rises in plasma noradrenaline concentrations. Conversely, reduction of sympathetic activity by drugs, nerve section or neurological disease resulted diminished the plasma levels of noradrenaline.
8 Although it cannot be doubted that manoeuvres that alter
sympathetic activity are associated with changes of plasma noradrenaline in the appropriate direction, it is by no means
certain that the results of these rather disparate sympathetic
stimuli can be extrapolated to a general proposition that plasma noradrenaline levels are a valid and quantitative marker
of sympathetic activity. There are three main areas of difficulty.
First, the plasma noradrenaline level is not determined
solely by the rate of release from sympathetic neurones. The
other major determinant is the plasma metabolic clearance rate,
which may vary several-fold from one individual to the next
without any apparent relationship to prevailing sympathetic
activity. Thus, for any given input of noradrenaline into the
circulation, a high clearance rate will result in a low plasma
level, while a low clearance rate produces a high plasma level.
Other factors which may also influence noradrenaline kinetics
include the rate of diffusion from synapse to plasma, the volume
of distribution, tissue binding, and uptake by platelets.
Second, any increase in sympathetic activity in
hypertensive patients would presumably be modest (but
sustained). Plasma noradrenaline levels measured in
hypertensive patients during supine rest are only 20-40% greater
than in normotensive controls. By contrast, the experimental
stimuli used to provoke noradrenaline release are often
relatively intense all-or-nothing stimuli which involve changes
9 in posture or central haemodynamics; plasma noradrenaline levels may rise by twofold or more. These methods are not well suited to assessment of small changes in sympathetic activity. Thus, evidence that small changes in sympathetic activity can be detected by small increases in plasma noradrenaline is lacking.
Third, the data concerning increases in plasma noradrenaline are based on acute studies. It is not known to what extent such increases would persist if the increase in sympathetic acivity were sustained, particularly since other parts of the sympathetic nervous system are capable of adapting to long-term changes in activity, eg receptor regulation.
In conclusion, the extent to which plasma noradrenaline adequately reflects sympathetic activity, and particularly small changes in sympathetic activity, is not well defined.
10 1.2 AIMS OF THIS STUDY
The overall purpose of the work reported in his thesis was to assess the usefulness, and to define the limitations of plasma noradrenaline levels as a biochemical index of sympathetic nervous system activity.
1.2.1 Assessment of generalised sympathetic activity.
The data is presented in two parts. The first part concerns the assessment of peripheral venous plasma noradrenaline in relation to 'generalised' sympathetic activity.
Specific aims were as follows.
1. Several related studies were performed to determine whether plasma noradrenaline levels were sensitive to small
changes in sympathetic activity.
2. A further aim was to determine over a broader range of
sympathetic activity whether a quantitative relationship existed
between changes in sympathetic activity and in plasma
noradrenaline.
3. It was also felt useful to consider the question from the
opposite side, namely whether a given increase in plasma
noradrenaline was always indicative of the same degree of change
in sympathetic activity.
11 1.2.2 Assessment of cardiac sympathetic activity.
The second part of the data concerns the estimation of sympathetic nervous activity in the heart. This was undertaken because the assessment of regional sympathetic activity using peripheral plasma noradrenaline is necessarily limited.
1. The initial aim therefore was to develop a method for estimation of intracardiac noradrenaline kinetics, since cardiac sympathetic activity is best described in terms of the rate of release of noradrenaline.
2. This method was then applied to the measurement of intracardiac noradrenaline turnover in man, at rest. The effects of sympathetic stimuli, and of propranolol were also assessed.
12 1.3 LAYOUT OF THE THESIS
A review of the relevant literature is presented in the next three chapters. Chapter 2 summarises the structure and function of the sympathetic nervous system; chapter 3 describes the role of the sympathetic nervous system in experimental and human hypertension; chapter 4 reviews the current evidence concerning the relationship between sympathetic nervous activity and plasma noradrenaline levels.
Chapters 5 to 9 concern 'noninvasive' studies on the role of peripheral venous plasma noradrenaline in the assessment of generalised sympathetic activity. Chapter 5 details the methodology used in these studies, with particular reference to the catecholamine assay technique; specific experiments and data are reported in chapters 6,1,8 and 5.
Chapters 10 to 12 are devoted to studies on the estimation of cardiac sympathetic activity. The theoretical basis for calculating intracardiac noradrenaline turnover is described in chapter 10. Chapters 11 and 12 report experimental data.
Finally, the conclusions drawn from this work are given in
chapter 13.
13 CHAPTER 2
THE STRUCTURE AND FUNCTION OF THE SYMPATHETIC NERVOUS SYSTEM
14 2.1 HISTORICAL INTRODUCTION
2.1.1 Autonomic nervous system
The autonomic nervous system comprises those nervous elements whose function in the body is to regulate the activity of the cardiovascular system, the exocrine glandular ducts, the pilomotor muscles of the skin, and those structures in the eye which contain smooth muscle such as the ciliary muscle and the pupil. In addition, the autonomic nervous system innervates the involuntary smooth muscle which is largely present in the hollow viscera such as the gastrointestinal and respiratory tracts.
As its name implies, the autonomic nervous system functions largely independently of conscious thought. This is a valuable evolutionary asset for higher animals such as man, since it allows the maintenance of the internal bodily environment by controlling vital functions such as tissue perfusion and temperature regulation without involving or distracting the higher cerebral centres. The term 'homeostasis' was coined by Cannon (1929) to describe this ability to maintain constancy of the internal bodily environment.
It was Galen, the Greek physician of the second century AD who first described the structures which we now know to be the nerve trunks and ganglia of the autonomic nervous system. He believed that the functioning of the body required consent or
'sympathy' between its different parts, and thought that this
15 was brought about by the passage of animal spirits through the nerves which he considered to be hollow tubes. Galen regarded the vagi and sympathetic trunks as a single unit, a dogma that was unchanged until 1545, when Estienne recognised that they were anatomically and functionally distinct structures. In
1751, Whytt made the first reference to the reflex nature of autonomic activity when he suggested that involuntary movements of the pupil and the bowel were initiated by local stimulation caused by nerve irritation. Later, in 1765, he also suggested that all 'sympathy' must be referred to the central nervous system since it occurred between parts of the body whose nerves were clearly not connected with each other. However, it was the
Danish anatomist Winslow who first used the term 'sympathetic' to describe a particular autonomic nerve.
The emergence of histological and physiological techniques in the nineteenth century allowed further under standing of the autonomic nervous system to be established on a clearer
anatomical and functional basis. The anatomists Ehrenberg,
Meissner and Auerbach respectively described cell bodies in
sympathetic ganglia, the submucous plexus of the intestinal tact, and the myenteric plexus. In 1851, Claude Bernard
demonstrated the vasoconsrictor function of some sympathetic
nerves.
The modern view of the autonomic nervous system is based
largely on the work of the anatomist WH Gaskell, and the
physiologist WN Langley. We now recognise that the autonomic
16 nervous system consists of sympathetic and parasympathetic divisions whose functions in the organs which they innervate are in general mutually antagonistic. This division was originally based on the anatomical studies of Gaskell (1916), and was later substantiated by the physiological and pharmacological work of
Elliot (1904), Dale (1914), Loewi (1921) and others.
2.1.2 The concept of neurohumoral transmission
These classic experiments demonstrated that the spread of electrical excitation from nerve to muscle depended on the release of a chemical 'neurotransmitter' at the nerve-muscle junction, and also that the neurotransmitters released by the sympathetic and parasympathetic divisions were different. The earliest evidence to support the neurohumoral concept was noted in 1904 by Elliot, who made two crucial observations concerning the actions of adrenaline. Firstly, he showed that adrenaline had no effects in tissues which were not sympathetically innervated; and secondly, that it retained its activity in sympathetically innervated tissues even after the sympathetic nerve supply to that tissue had been allowed to degenerate.
Since the latter observation was inconsistent with Langley's earlier hypothesis that adrenaline acted by stimulating sympathetic nerve endings, Elliot suggested that adrenaline itself might be the chemical stmulant released from sympathetic nerve endings. In 1906, Langley further developed the concept that nerve-to-muscle actvation depended upon the release of a chemical stmulant from nerve endings, and suggested that these
17 stimulants might be acting on 'receptive substances' located on specific areas of the muscle.
2.1.3 Parasympathetic neurotransmission
Somewhat earlier, Schmiedeberg and Koppe (1868) had reported that the effects obtained by injection of the extract
(named 'muscarine') of the toadstool Amanita muscaria were similar to the effects produced by electrical stimulation of the vagus. In both cases, the effects were abolished by atropine.
Susequently, Dixon, in 1907, described a series of experiments
he in which claimed successful extraction of a muscarine-like A substance from animal hearts following prolonged vagal stimulation, but the idea of parasympathetic neurotransmission did not gain widespread acceptance until the elegant and classic experiments of Otto Loewi in 1921. Loewi stmulated the vagus nerve of a frog heart in situ, reducing the rate and force of the spontaneous cardiac contractions. He then showed that the perfusate from this heart would have similar effects when pipetted onto a second, spontaneously beating frog heart. He suggested that the inhibitory effects were caused by the release into the perfusate of a chemical substance which he termed
'Vagusstoff', later identified as acetylcholine by Feldberg,
Krayer and others.
Subsequently, Dale (1914) discovered that in addition to the muscarine-like effects produced by low doses of acetylcholine, high doses of acetylcholine had very different
18 effects which were similar to the actions of the alkaloid nicotine, and which were not abolished by atropine. It was later shown that muscarinic effects predominated at the neuromuscular junctions, while the nicotinic effects predominated in autonomic ganglia.
2.1.4 Sympathetic neurotransmission
Folowing the work of Elliot and of Langley, Barger and
Dale (1910) examined the effects of a series of compounds related to adrenaline and reported that one of them, called noradrenaline, had reproduced more closely the effects of sympathetic stimulation in some tissues than had adrenaline.
Dale interpreted this as evidence against neurotransmission since it was inconsistent with Elliot's earlier data which suggested that it was adrenaline released at sympathetic nerve endings. At that time, noradrenaline was not known to occur naturally in the body. These discrepancies were not finally resolved until 1946 when improved chemical techniques allowed von Euler to show that noradrenaline is the sympathetic neurotransmitter in man and in higher animals. However, by that time, Dale (1933) had already proposed the terms 'cholinergic' and 'adrenergic' to distinguish the actions and effects of the nerves from the two divisions of the autonomic nervous system.
(The term 'adrenergic' reflecting the contemporary belief that adrenaline was the likely transmitter at sympathetic nerves).
These terms remain in use today (Day, 1979).
19 2.2 GENERAL ANATOMY
The sympathetic nervous system is made up of cellular elements distinguished by their embyological derivation from neural crest tissue, and by their distinct biochemical and pharmacological properties which relate to the synthesis and release of the postsynaptic neurotransmitter catecholamines.
Sympathetic neural tissue is distributed both centrally within the brain and spinal cord, and peripherally.
2.2.1 Central sympathetic connections
The main centres of the brain concerned with the regulation of sympathetic (and parasympathetic) activity are situated in the hypothalamus and medulla oblongata. Tracts of efferent fibres leave the hypothalamus and synapse both with the medullary centres, and also in the spinal cord with the cell bodies of the sympathetic neurones located in the lateral horns of the grey matter. The medullary centres are the main sites of vasomotor integration, and also receive afferent fibres from the limbic system and from higher cerebral centres.
2.2.2 Peripheral connections
Fibres from the cerebral centres pass down the spinal cord to synapse at levels T1 to L3 of the thoracolumbar cord.
Preganglionic sympathetic fibres pass from the spinal cord and
20 emerge from the spinal column with the segmental spinal nerves of Tl to L3. They then branch off from these spinal nerves as myelinated trunks (white rami communicantes) and join the paravertebral sympathetic ganglia.
The preganglionic fibres connect with the postganglionic effector neurones in two main ways. Firstly, the preganglionic fibres may synapse with the cell body of the postganglionic neurone within the paravertebral sympathetic chain (although it may first pass up or down the chain to synapse at a different level from which it entered). The postganglionic fibre then emerges from the paravertebral chain either as a sympathetic nerve trunk, or returns to its original spinal nerve (in the grey rami communicantes) and ultimately branches off the somatic nerve to pass to the effector organs. Secondly, the preganglionic fibre may pass right through the paravertebral chain without synapsing, and synapse instead with peripheral sympathetic ganglia that lie in the abdominal cavity, eg the coeliac and mesenteric ganglia. Finally, a few fibres pass directly from the cord to the adrenal medulla, which they innervate (Day, 1979).
2.2.4 Cardiac sympathetic supply
Cardiac sympathetic fibres originate in the
intermediolateral columnsof the upper five or six thoracic
segments of the spinalcord and synapse with postganglionic
neurones of the corresponding paravertebral sympathetic ganglia
21 and in the cervical ganglia. The postganglionic fibres travel to the heart as the (left and right) superior, middle and
inferior cardiac nerves, respectively, from the (left and right)
superior, middle and inferior cervical ganglia. A few postganglionic fibres pass directly to the heart from the upper
five or six thoracic ganglia, approaching the base of the heart
along the adventitial surfaces of the great vessels. These are
then distributed to the various chambers as an extensive
epicardal plexus which penetrates the myocardium, usually
accompanying the branches of the coronary vessels. The atria
are about three times more densely innervated than the
ventricles. This is illustrated in Figure 2-1.
2.2.5 Innervation of peripheral vessels
Postganglionic sympathetic fibres from the grey rami, or
from the splanchnic and cervical ganglia join the large arteries
and accompany them as an investing network, of fibres to the
resistance vessels (arterioles) and capacitance vessels.
Sympathetic nerve fibres differ from the motor fibres of
the somatic nerves, which range from about 1 to 20 microns in
diameter and are termed A fibres. The sympathetic preganglionic
fibres are also myelinated, but are about 3 microns in diameter
and are termed B fibres. The postganglionic fibres are termed C
fibres and are about 1 micron in diameter. Since the velocity
of nerve conduction is inversely related to diameter; thus the
conduction velocity in the sympathetic B and C fibres (5-20 m/s)
22 Superior cervical ganglion
Superior cardiac nerve
Middle cervical ganglion Middle cardiac nerve Inferior cardiac nerve Stellate Inferior cervical ganglionganglion T1
Paravertebral sympathetic chain
Figure 2-1. The sympathetic innervation of the heart. is slower than in the somatic A fibres (100-120 m/s).
The structure of sympathetic nerve terminals has been studied with fluorescent histochemical techniques (Falck et al, 1962) . In blood vessels, the terminal portions of the postganglionic fibres undergo repeated divisions to form a dense network of fibres, the 'terminal reticulum'. (This may include parasympathetic cholinergic fibres and visceral afferent fibres according to the tissue.) In blood vessels, the terminal reticulum runs in the adventitial layer; sympathetic axons show numerous terminal varicosities at the sites of neuroeffector synapses, and a single axon synapses at frequent intervals along its length at the area of apposition (Hillarp, 1957) .
2.3 SYNTHESIS AND RELEASE OF NORADRENALINE FROM SYMPATHETIC
NERVE TERMINALS
2.3.1 Biosynthesis of catecholamines
The sequence of catecholamine biosynthesis shown in Figure
2-2 was proposed by Blashko, based on studies of adrenal medullary chromaffin tissue. However, the presence of the
synthetic enzymes required for the pathway has been demonstrated
in all sympathetic tissues, including the brain, sympathetic
ganglia, and sympathetic postganglionic nerves. In man,
phenylethanolamine N-methyltransferase (PNMT) is present only in
brain and adrenal medulla; it is absent in sympathetic neurones
24 CATECHOLAMINE BIOSYNTHESIS
tyrosine HO c h 2- c h - n h 2 I COOH
tyrosine hyroxylase ▼
DO PA „ 7 CH2-CH-NH2 I COOH
DOPA decarboxylase ▼ HO
dopamine HO ( -c h 2- c h 2- n h 2
dopamine fi hydroxylase
▼ HO
noradrenaline HO ( c h - c h 2- n h 2 I OH
phenylethanolamine N-methyltransferase
HO
adrenaline HO ( c h - c h 2- n h - c h 3
Figure 2-2 The biosynthesis of catecholamines.
25 where the transm itter is noradrenaline (Blashko, 1973) .
Histofluorescense and fractionation studies of sympathetic neurones indicate that the main site of catecholamine synthesis is peripheral: in the axon, at, or near the nerve endings
(Geffen and Livett, 1971). Tyrosine is taken up by the cell from the extracellular fluid and converted to L-DOPA by tyrosine hydroxylase which is free in the cytosol, this being a rate-limiting step. The conversion of L-DOPA to dopamine also occurs in the cytosol, but the formation of noradrenaline first requires the active uptake of dopamine into the storage vesicles where it is a substrate for the enzyme dopamine beta hydroxylase, which is largely membrane-bound (Hortnaegl et al,
1972). The storage vesicles subsequently migrate down the axon to the terminal varicosities (Smith 1979).
2.3.2 Neuronal uptake of noradrenaline
The cell membrane of the sympathetic nerve terminal takes up noradrenaline present in the extracellular fluid and synaptic cleft, whether the noradrenaline has been liberated by the nerve terminal itself, or whether it has diffused from the bloodstream. The process of neuronal uptake ("uptake 1") results from the activity of a membrane carrier system that requires metabolic energy. The uptake mechanism is saturable and has a very high affinity for noradrenaline; it is also stereoselective, and the affinity for the naturally-occurring
1-isomer is 5 times greater than for the d-isomer. Like most
26 other active membrane transport systems, neuronal uptake is temperature selective, and is inhibited by metabolic poisons and by anoxia. The active carrier is linked to the Na-K ATPase transport system, and conditions that affect the activity of that enzyme also affect neuronal uptake. A variety of other amines related to noradrenaline can also act as substrates for the neuronal uptake system. The affinity of adrenaline is approximately half that of noradrenaline in mammalian tissues; dopamine is similar. Other high-affinity substrates include methylnoradrenaline, alpha methyladrenaline, tyramine, octopamine and alpha methyltyramine (Burgen and Iversen, 1965;
Iversen, 1967).
2.3.3 Intraneuronal storage of noradrenaline
Part of the noradrenaline taken up by the neuronal uptake process is oxidised to 3,4-dihydroxyphenylglycol (DOPEG) by monoamine oxidase in the cytosol before it reaches the storage sites. The majority of noradrenaline (whether taken up or synthesised) is stored in the vesicles of the terminal varicosities as a complex with proteins, notably with
Chromogranin A. Fractionation and ultrastructural studies show two types of vesicles, 'small' and 'large', and their features are shown in table 2.1.
27 Table 2.1 Properties of noradrenergic vesicles
diameter constituents
Large vesicle: 85 nm noradrenaline
adenosine triphosphate
dopamine beta hydroxylase
- 80% membrane-bound
- 20% soluble
Chromogranin A
other chromogranins
Small vesicle: 50 nm noradrenaline
dopamine beta hydroxylase
? adenosine triphosphate
? other proteins
The large vesicles appear to be formed in the cell bodies of sympathetic neurones, possibly by budding from the endoplasmic reticulum. At this point their noradrenaline content is low, but this increases progressively (presumably due to synthesis) as the large vesicles are transported down the axon to the nerve terminals. Both large and small vesicles are present in the terminals; the latter, which also have a high noradrenaline content, are not found elsewhere. The precise roles of two types of vesicles in noradrenaline release is still
28 unclear, as is the origin of the small vesicles themselves
(Smith, 1979; Fillenz, 1979).
Small vesicles may be formed in situ from smooth endoplasmic reticulum, but current evidence favours formation as out-buddings ("coated pits") from empty large vesicles which have discharged their contents by exocytosis. The small vesicles are initially empty, but appear to be able to synthesise or take up proteins and noradrenaline, following which they also have a role in the exocytotic release of noradrenaline (Smith, 1979).
2.4 NORADRENALINE RELEASE
2.4.1 Nonexocytotic release
Part of the noradrenaline contained in the neuronal storage vesicles continuously diffuses out into the cytosol and from there to the extracellular space. However most is deaminated by cytoplasmic monoamine oxidase before it reaches the synaptic cleft, so that no response occurs. In the rat heart, this leakage is accelerated under conditions of anoxia, ischaemia and energy depletion. There is evidence suggesting involvement of the Na-K ATPase - linked neuronal uptake process operating in the reverse direction, i.e. outward transport
(Schoemig et al 1987).
29 Noradrenaline can also be released pharmacologically through displacement from its storage sites by sympathomimetic compounds of which amphetamine and tyramine are examples. A increase in the neuroplasmic calcium ion concentration is not required for this process (Smith, 1973).
2.4.2 Exocytotic release
Release of noradrenaline from the terminal varicosities is initiated by the arrival of action potentials that originated in cell bodies. Depolarisation of the terminals causes a rise in the membrane conductance of sodium and calcium ions by activation of specific channels; both ions then enter the terminal passively down concentration gradients. This increase in the intracellular calcium ion concentration causes migration of the storage vesicles towards the neuronal membrane, possibly transported via a calcium-activated microfilament or contractile protein. The vesicles then fuse with the plasma membrane, and can discharge some, or all, of their contents into the synaptic cleft.
Exocytotic release can also be initiated experimentally by electric current, by membrane depolarisation due to increases or decreases in the postassium ion concentration, by nicotine, and by calcium ionophores (Blaustein, 1979; Vanhoutte, 1980).
2.4.3 Presynaptic modulation of exocytotic release
30 There is now good evidence that noradrenaline release can be modulated at the presynaptic level by endogenous catecholamines and by other neurohumoral compounds. This is mediated by facilitatory and inhibitory receptors located presynaptically on the neuronal membrane of the terminal varicosities. Alpha-adrenoceptors appear to have inhibitory effects on the stimulation-evoked release of noradrenaline from sympathetic nerve endings; this mechanism is activated when higher concentrations of noradrenaline are reached in the synaptic cleft. Further release of transmitter is thereby inhibited. Binding studies with ligands selective for alpha-1 and alpha-2 adrenoceptors suggest that the presynaptic receptors are principally of the alpha-2 subtype.
The main lines of evidence can be summarised as follows.
(1) The stimulation-evoked overflow noradrenaline from various tissues is reduced in the presence of alpha-2 agonists, whereas alpha-2 antagonists increase transmitter overflow. (2)
This regulation is independent of whether the postsynaptic receptors of the effector organ are alpha or beta. (3) This increase in transmitter release is not affected by atrophy of the postsynaptic tissues, and (4) is also observed in cultured nerve endings where there are no postsynaptic receptor cells present. (5) degeneration of noradrenergic nerve endings
following chemical sympathetomy with 6-hydroxydopamine leads to
a reduction of specific alpha-2 ligand binding (in tissues
devoid of postsynaptic alpha-2 receptors).
31 Facilitation of stimulation-evoked noradrenaline release appears to be mediated by beta-2 receptors. Specific beta-2 agonists enhance transmitter release, while beta-1 agonists are without effect. These effects are blocked by beta-2 antagonists, and by non-specific (beta-1 plus beta-2) antagonists, but not by antagonists which are specific for beta-1 receptors only (Langer, 1979, 1981).
A number of other presynaptic receptors have been identified. Muscarinic receptors have been shown to inhibit transmitter release in vitro and in vivo (Levy and Blattberg,
1976; Vanhoutte, 1976) . In organs with both adrenergic and cholinergic innervation, stimulation of the parasympathetic nerve trunks decreased the overflow of noradrenaline caused by
simutaneous stimulation of the sympathetic innervation(Mathe et
al, 1977; Muscholl, 1979).
Prostaglandin receptors of the PGE1 and PGE2 series
inhibit noradrenaline release in several species and tissues, but the importance of this mechanism is not yet certain
(Stjarne, 1979).
Inhibitory dopamine and opiate receptors have also been
described in peripheral noradrenergic nerve endings.
Presynaptic facilitatory angiotensin II receptors have also been
reported in peripheral nerve endings, including the heart
(Langer, 1981).
32 2.5 SYNAPTIC CLEFT
2.5.1 Synaptic cleft width
The terminal varicosities of sympathetic nerves are usually enveloped by the Schwann cells in such a way that a bare area of the varicosity is left orientated towards the nearest effector cells. The gap between them is termed the synaptic cleft, and this varies greatly in width. In general, the smaller the blood vessel, the narrower the cleft. The mean synaptic cleft widths documented in animal studies include rabbit pulmonary artery (1,900 nm) , rabbit ear artery (500 nm) , rat mesenteric arteries (200-500 nm), guinea-pig uterine arteries (200 nm), rat portal vein (100 nm), cat nictitating membrane (25 nm) and guinea-pig vas deferens (20 nm) (Bevan
1979) .
2.5.2 Functional consequences
Noradrenaline concentrations within and without the synaptic cleft have been estimated in vascular tissues with narrow and wide synaptic clefts by comparing the responses to neurogenic stimulation and to exogenously-applied noradrenaline concentrations. In wide clefts the intra and extrasynaptic noradrenaline concentrations were similar at about 10-7M; by
contrast, in narrow clefts the intrasynaptic concentrations
33 rose as high as 10-5M and the extrasynaptic concentration of* was m the 10-9M range. This existence^a noradrenaline concentration gradient between the inside and outside of small clefts suggests the leakage of a transmitter is restricted, and that the neuronal uptake of the release transmitter is proportionately greater in narrower than in wider clefts. The converse also holds, namely that the uptake of exogenous noradrenaline by narrow synaptic clefts is less than by large synaptic clefts, presumably because ingress of the transmitter is restricted. These observations also imply that both release and neuronal reuptake of noradrenaline is restricted to the bare area of the nerve terminal, i.e. to the presynaptic membrane
(Bevan and Su, 1974; Bevan, 1977). Furthermore, the alpha-receptor-mediated from presynaptic inhibition seems to be more important in small synapses (Bevan, 1979).
2.6 PRESYNAPTIC RECEPTORS
2.6.1 Introduction
The responses of the postsynaptic effector cell are initiated by the binding of the transmitter to specific lipoprotein receptors. The localisation of the receptors may be both intra and extrasynaptic, mediating responses to neurogenic stimulation and to circulating catecholamines, respectively.
The magnitude of the response depends principally on the concentration of the transmitter, the affinity of the receptor,
34 and the intrinsic activity of the effector tissue.
Classification of the major receptor subtypes has been based on differences in the rank order of potency exhibited by a range of agonists in tissues where one or other type of receptor predominates.
In 1948, Ahlquist noted that a series of six sympathomimetic amines had one order of potency when producing vasoconstriction, excitation of the uterus and ureter, contraction of the nictitating membrane, dilation of the pupil and inhibition of the gut. The same series of amines had an entirely different order of potency in myocardial stimulation, inhibition of the uterus and vasodilatation. Ahlquist (1948) concluded that there were two distinct types of adrenergic receptor, which he termed alpha and beta. Further differentiation of beta-receptor subtypes in beta-1 and beta-2 was first reported in 1967 by Lands, Ludena and Buzzo. This classification is again based on finding that the rank order of potency of several agonists fell into two main categories when tested on a series of tissues. Verification of the two beta-receptor subtypes has since been achieved by a synthesis of more selective compounds, and radioligand binding studies.
The differentiation of the alpha-receptor subtypes in alpha-1 and alpha-2 was demonstrated by Starke et al (1975), during the course of studies on presynaptic receptors, phenylephrine and methoxamine preferentially stimulated postsynaptic alpha-receptors - these were termed alpha-1.
35 oxymethoxamine and clonidine however stimulated inhibitory presynaptic alpha-receptors, and these were termed alpha-2.
Initially, alpha-2 receptors were thought to be located exclusively on the presynaptic membrane, but it has recently been shown that they are also located extrasynaptically on vascular smooth muscle cells. Here, they appear to respond preferentially to circulating catecholamines rather than to neuronal noradrenaline (Jie et al, 1985; Elliott and Reid,
1983).
2.6.2 Postsynaptic receptor and tissue responses
The distribution of postsynaptic alpha and beta receptors and their subtypes in cardiovascular tissues, and the responses of these tissues to receptor stimulation, are summarised in table 2.2.
2.7 REMOVAL OF NORADRENALINE
2.7.1 Introduction
The fate of released noradrenaline is complex. Receptor stimulation in synaptic cleft is terminated mainly by neuronal reuptake, but a small proportion of the released noradrenaline cleft may be taken up outside the synaptic^by extraneuronal mechanisms
('uptake-2') which are located mainly in smooth muscle.
36 Table 2.2 Distribution of adrenergic receptors in cardiovascular tissues, and their postsynaptic effects.
______postsynaptic effects______organ/tissue alpha receptor (type)____ beta receptor (type)
heart. SA node - beta-1: increased HR++ beta-2: HR++
atria “ beta-1: increased con tractility, & conduct ion velocity++.
AV node & beta-1: increased auto His-Purkinje maticity & conduction system ' velocity**.
ventricles beta-1: increased con- tractili ty, conduction velocity and automat- icity+++
arterioles. coronary alpha-1: const rict ion^- beta-2: dilatation**
skin/mucosa alpha-1: constriction+H- -
skeletal muse alpha-1: constriction++ beta-2: dilatation** alpha-2: constriction++
cerebral alpha-1: constriction* -
pulmonary alpha-1: constriction* beta-2: dilatation+
viscera alpha-1: constriction++ beta-2: dilatation*
renal alpha-1: constriction++ beta-1: dilatation* beta-2: dilatation*
veins. alpha-1: constriction++ beta-2: dilatation++
kidney. - beta-1: renin secretion++
Abbreviations. SA, sinoatrial. AV, atrioventricular. Muse, muscle. Magnitude of responses: + slight; + moderate; ++ marked.
37 Noradrenaline is then metabolised by the enzymes monoamine oxidase and catechol-O-methyltransferase.
2.7.2 Neuronal re-uptake
The actions of the released noradrenaline are terminated chiefly by the neuronal uptake mechanism, which has already been described in Section 2.3.2. Overall, reuptake counts for approximately 80% of the released noradrenaline, and the remaining 20% diffuses out of the synaptic cleft into the plasma.
2.7.3 Extraneuronal uptake
Stromblad and Nickerson first noted in 1961 that tissue components other than nervous tissue had the capacity to retain noradrenaline by a process of extraneuronal uptake. Fluorescence histochemical techniques have demonstrated that a variety of tissues take up catecholamines from the extracellular fluid; these include cardiac muscle, vascular smooth muscle - especially arterial - collagen, and elastic. The properties of the extraneuronal uptake mechanisms are not fully defined, and indeed uptake 2 mechanisms have been described in a heterogeneous group of cells in which the properties of the uptake mechanism are not always identical.
In smooth muscle, noradrenaline is located mainly at intracellular sites throughout the cytoplasm and concentrated on
38 the nucleus. There are also noradrenaline binding sites at the periphery of the cell on the basement membrane. Intracellular uptake is concentration-dependent and appears to obey
Michaelis-Menten kinetics. There is some requirement for sodium ions, but the relationship to metabolic energy requirements is not clear. The affinity of uptake-2 is low compared to neuronal uptake. The degree of accumulation of the amine is modest, and the ratio of the tissue to medium concentration does not exceed
3. However, the total quantity of catecholamine retained can be considerable because of the bulk of tissue involved.
Uptake 2 is inhibited by various drugs including some alpha-blockers - particularly phenoxybenzamine - and corticosteroids. The mechanism is not stereoselective.
Inhibitors of neuronal reuptake are ineffective.
The physiological role of extraneuronal uptake has been difficult to establish with certainty. It is not directly involved in the termination of transmitter in neuronal release, but may have a role in the uptake and metabolism of circulating catecholamines (Gillespie 1973).
2.7.4 Metabolism of noradrenaline
Noradrenaline is metabolised chiefly by the enzymes monoamine oxidase (MAO) and catechol-O-methyltransferase (COMT).
The major metabolic pathwys are shown in figure 2-3. Deamination
by MAO appeares to be the chief route of meatabolism in the
39 HO- -CHOH [MAO] HO -CHOH HO CH., HO HCO NH, Norepinephrine
NH, Normetanephrine
Figure 2-3. Steps in the metabolism of noradrenaline.
Terms and abbrevations. Norepinephrine, noradrenaline.
DOPGAL, 3,4-dihydroxyphenylglycoaldehyde. DOPEG (or DHPG),
3,4-dihydroxyphenylglycol. DOMA, 3,4-dihydroxymandelic acid.
MOPEG (or MHPG), 3-methoxy-4-hydroxyphenylglycol. VMA,
3-methoxy-4-hydroxymandelic acid. MOPGAL, 3-methoxy-4-hydroxy
-phenylglycoaldehyde. COMT, catechol-O-methyltransferase.
MAO, monoamine oxidase. ALD RED, aldehyde reductase. ALD
DEHYD, aldehyde dehydrogenase.
40 neurone. Mitochondrial MAO is responsible for the breakdown of extragranular stores of noradrenaline in the nerve terminal, including noradrenaline taken up by the uptake-1 mechanism. In tracer studies, 3,4-dihydroxyphenylglycol is the major compound which accounts for the increment in the radiolabel when the nerves are stimulated. Intact noadrenaline accounts for less than a third.
COMT is present in nearly all tissues, in association with the cofactor S-adenosyl methionine which is a methyl group donor. O-methylation is the major route of meatabolism in extraneuronal tissues; this is the main metabolic pathway for circulating catecholamines. In the periphery, the main end product of endogenous catecolamine metaboilsm is 3-methoxy,
4-hydroxymandelic acid (vanillylmandelic acid: VMA) which accounts for 60% of urinary catecholamine metabolites. 30% are in the form of 3-methoxy-4-hydroxyphenylglycol (MOPEG or MHPG).
Unchanged adrenaline and noradrenaline comprises less than 1%
(Kopin, 1979).
41 CHAPTER 3
THE ROLE OF THE SYMPATHETIC NERVOUS SYSTEM IN HYPERTENSION
42 3.1 INTRODUCTION
Because autonomic nervous tone, and sympathetic nervous activity in particular, are the dominant mechanisms for effecting acute changes in blood pressure and heart rate, it has long been a matter of speculation that sustained changes in sympathetic activity might be an important factor in the initiation or maintenance of essential hypertension in some patients. This concept of 'neurogenic' hypertension stems from haemodynamic and biochemical observations made over the past twenty years. In this chapter, the evidence concerning the role of the sympathetic nervous system in human and experimental hypertension will be reviewed.
3.2 HAEMODYNAMIC STUDIES IN MAN
3.2.1 Haemodynamic features of essential hypertension.
Whereas established hypertensives are known to have elevated peripheral resistance and a normal or reduced cardiac
output (Freis, 1960), observations made in the early stages of
hypertension, (variously termed 'labile', 'borderline' or
'mild') showed that a proportion of patients had elevated blood pressure due to increased cardiac output in the presence of
normal peripheral resistance. Such findings have been
documented in detail in children, adolescents and as well as in
adults under the age of 40 years. (Bello et al, 1965;
43 Lund-Johansen, 1967; Kuramoto et al, 1968; Frohlich et al, 1969;
Safar et al, 1970; Julius et al, 1971; Davignon et al, 1977;
Dunstan and Tarazi, 1977.). These features are generally accompanied by increased heart rates and oxygen consumption
(Julius and Conway, 1968). The haemodynamic features are summarised in table 3.1.
However, not all hypertensives in this age group share the same features, and cardiac outputs may be normal or low as well as high (Messerli et al, 1968). The high-output group probably represents an important sub-group. Longitudinal studies suggest that the haemodynamic mechanism of hypertension changes with the passage of time: cardiac output falls and peripheral vascular resistance rises progressively to approach the low flow-high resistance pattern found in older hypertensives (Lund-Johansen,
1980).
3.2.4 Autonomic 'imbalance' in some hypertensives.
The haemodynamic pattern of these patients - tachycardia,
increased cardiac output and raised oxygen consumption -
resemble the effects of sympathetic stimulation. Julius and
colleagues have investigated the haemodynamics in borderline
hypertensives and assessed the effects of autonomic blockade.
Cardiac output and heart rate was raised in the hypertensive
patients at rest. After propranolol, the cardiac outputs in both
groups were lower and the differences between them diminished.
After complete autonomic blockade with propranolol and atropine,
44 Table 3.1 Haemodynamic features of subjects with 'mild' or
'labile' hypertension during supine rest.
MAP Cl TPR Index HR Stroke Index
Study (mmHg) (L/min/m2) (KPa/L/m2) (bpm) (ml/beat/m2)
(a) H 108 4.1 211 81 51
C 91 3.3 221 74 44
(b) H 102 4.1 197 75 56
C 84 3.4 195 63 55
(c) H 106 3.5 240 77 46
C 93 3.0 244 68 45
(d) H 105 4.1 210 80 51
C 86 3.1 222 72 44
(e) H 100 3.8 222 76 50
C 83 3.3 209 67 50
Abbreviations: MAP, mean arterial pressure. Cl, cardiac index
TPR, total peripheral resistance. HR, heart rate. C, control.
H, hypertensive.
Studies cited: (a) Bello et al, 1965. (b) Lund-Johansen, 1967
(c) Frohlich et al, 1969. (d) Safar et al, 1977. (e) Julius
et al, 1971.
45 cardiac output in the hypertensives was lower than controls and the heart rates became similar at about 100 bpm. This suggested
that (a) the instrinsic activity of the sinus node was normal in the hypertensives, and (b) that their increased resting cardiac
output was maintained by autonomic imbalance (Julius et al,
1971, Julius and Esler, 1975) . Baroreflex sensitivity is diminished in these patients, but normalises with neostigmine
(Trimarco et al, 1981). Similarly, salivary flow, which is
cheifly under parasympathetic control, is reduced (Henquet et
al, 1982) . Thus there is evidence that a disturbance in
autonomic function characterised by excessive sympathetic
activity and reduced parasympathetic activity may be present in
these patients.
3.3 ROLE OF THE SYMPATHETIC NERVOUS SYSTEM IN EXPERIMENTAL
ANIMAL HYPERTENSION
3.3.1 Introduction
A number of animal models have been developed to study the
various mechanisms of hypertension, and most of these mimic
specific aetiological factors such as excessive sodium intake,
increased mineralocorticoid secretion, activation of the
renin-angiotensin system and alteration of baroreflex
sensitivity. Current evidence suggests that sympathetic
activity is increased in most forms of experimenal hypertension
in the rat (de Champlain et al, 1977).
46 3.3.2 DOCA-salt hypertension
In rats rendered hypertensive by the administration of corticosterone and saline (DOCA-salt hypertension), the retention and storage or noradrenaline in cardiac and vascular tissue is reduced, while the rate of noradrenaline synthesis and turnover is increased, indicating increased sympathetic activity compared with control rats (de Champlain et al, 1967; Krakoff et al, 1967). This was consistent with the more rapid urinary efflux of radioactive label after administration of tritiated noradrenaline (de Champlain et al, 1969) and subsequently by the finding of two or three-fold higher catecholamine levels in plasma. Conversely, chemical sympathectomy with
6-hydroxydopamine lowers the blood pressure, demonstrating that the functional integrity of the sympathetic nervous system is required for the development of DOCA-salt hypertension (de
Champlain et al, 1976).
3.3.3 Experimental renal hypertension
Cardiac turnover of noradrenaline is accelerated in animals with various forms of renal hypertension. In the one-clip renal hypertensive rat, combinations of chemical sympathectomy, adrenalectomy and nephrectomy were used to assess the relative contributions of the sympathetic nervous system, adrenal medulla and the renin-angiotensin system. The greatest fall in blood pressure was obtained after chemical sympathectomy
47 (de Champlain and van Ameringen, 1972). In the one-kidney model, plasma noradrenaline became raised progressively up to a month after arterial constriction when compared with controls
(Dargie et al, 1975). More recently it has been shown that the development of hypertension in young renally hypertensive rats requires the functional integrity of the renal sympathetic nerves.
3.3.4 Spontaneously hypertensive rats
Even though not all strains of genetically hypertensive rats showed persistent evidence oF increased sympathetic activity when fully grown, normal sympathetic function appears to be essential for the early development of the disease.
Adrenalectomy, immunosympathectomy and section of the renal sympathetic nerves inhibit the development of hypertension
(Ozaki, 1966; Smirk, 1970; Winternets et al, 1980) . Increased plasma and cardiac turnover has been demonstrated in young spontaneously hypertensive SHR rats. This strain also shows an exaggerated increase of plasma noradrenaline in response to salt loading (Yamori et al, 1973; Nagaoka and Lovenberg, 1976).
48 3.4 BIOCHEMICAL EVIDENCE OF INCREASED SYMPATHETIC ACTIVITY
IN HUMAN ESSENTIAL HYPERTENSION
3.4.1 Introduction
Although direct neurographic recordings of sympathetic nervous activity have been described by various groups (Sundlof et al, 1981), their use is confined to the more accessible superficial tissues, and it has not been possible to determine sympathetic activty in the majority of vascular beds. The assessment of sympathetic nervous activity is therefore generally indirect and may be achieved by biochemical measurement of the products of neuronal noradrenaline release.
Relevant compounds include adrenaline, noradrenaline, dopamine and their metabolites in plasma, urine and occasionally in cerebro-spinal fluid. Reliable assays capable of measuring very small concentations of these compounds have been available only relatively recently, since he mid 1970s. The enzyme dopamine beta hydroxylase has also been studied since it is released stoichiometrically with neuronally derived noradrenaline.
3.4.2 Serum Dopamine Beta Hydroxylase
Dopamine beta hydroxylase (DBH) has been investigated as a possible marker of sympathetic neuronal activity since it is
49 present in noradrenaline storage vesicles and is released stoichiometrically with noradrenaline (Axelrod, 1972) . Increased
DBH levels have been reported in myocardial infarction
(Gutteberg et al, 1976), and a positive correlation between catecholamine excretion and plasma DBH levels was reported by
Shauberg and Kirshner (1976). Some but not all studies suggested a relationship between DBH, blood pressure and sympathetic activity (Wetterberg et al, 1972; Horwitz et al,
1973; Geffen et al, 1973) but plasma noradrenaline and DBH levels do not always vary in the same direction in various forms of stress such as exercise or electric shock (Wetterberg et al,
1972).
Extremely wide variations in DBH levels had been shown between individuals whose levels of sympathetic activity assessed by other means seemed normal, and further studies by
Weinshilbaum et al (1975) showed the importance of genetic factors as the principal determinent of serum DBH levels. Thus, in some families, the occurrence of very low DBH levels appears to be detemined by an autosomal recessive gene, and not by the degree of sympathetic nervous activity. Although it is evident that DBH levels can increase with intense or prolonged sympathetic nervous system stimulation, the magnitude of these changes is small and relate poorly to changes in plasma noradrenaline. This is partly due to the great diffence in plasma half lives which is two to three minutes for noradenaline but several hours for DBH (Rush and Geffen, 1972). Furthermore the large molecular weight of the DBH molecule, approximately
50 300,000, presumably confers very different diffusion properties compared with noradrenaline. Thus, no consistent or convincing relationship between DBH and sympathetic nervous activity has emerged, and it is unlikely that serum DBH levels can be used to determine changes in sympathetic activity (Kopin, 1979) .
3.4.4 Plasma catecholamines in human essential hypertension.
Improved catecholamine assays became available from the early 1970s with development of a number of highly sensitive and precise isotopic radio enzymatic techniques (Engelman et al,
1968; Louis and Doyle, 1971; Henry et al, 1975; da Prada and
Zurcher, 1976). These replaced the earlier fluorimetric techniques.
In 1970 Engelman, Portnoy and Sjoerdsma reported that plasma catecholamine levels were strikingly raised in patients with essential hypertension, the difference in catecholamine levels being two-fold: 0.24+0.07 ng/ml in normotensives and
0.45+160 ng/ml in the hypertensives. This difference was highly significant at p<0.001. A similar result was soon reported by d'Quattro and Chan who found a plasma noradrenaline of 0.27+0.02 ng/ml in normotensives compared with 0.34+0.15 ng/ml in the hypertensives (p<0.05). The suggestion that these findings might reflect increased sympathetic activity in the hypertensive patients was supported by the findings of Louis, Doyle and
Anavekar (1973)who showed agood correlation between plasma
noradrenaline and diastolic blood pressure in the range of 70 -
51 140 mmHg. Furthermore, a close correlation was also found between the fall in blood pressure and the fall in plasma noradrenaline after ganglion blockade. Taken together, these findings suggested that not only was sympathetic activity generally involved as a contributory factor in essential hypertension, but that excessive sympathetic activity might be quantitatively related to raised blood pressure.
Following these key observations, a number of similar reports were published up to 1976. The majority of these showed significant differences in the plasma noradrenaline levels of hypertensive and normotensive patients and strengthened support for the aetiological role of excessive sympathetic activity in essential hypertension. These data are summarised on table 3.2.
3.4.5 The effects of age on plasma noradrenaline levels, and its relevance to the composition of control groups.
Publications by Lake et al and by Sever et al in 1977 were
the first to draw attention to the importance of the age and
composition of the control group in such studies. Lake et al
(1977) studied 67 hypertensive patients whose mean age was 44
years, and 84 normotensive controls whose mean age was 33 years.
When noradrenaline was plotted against age in the combined
groups there was considerable variability, but a significant
positive relationship emerged, showing that plasma noradrenaline
levels tended to rise with increasing age. Overall the mean
52 in o M CO o co CO CO O co O M O O co o
l/l t—4 o * * * * * * P CM co CM r-v VO in 53 plasma noradrenaline was higher in hypertensive patients than in normotensive patients but this difference disappeared when the data were corrected for age. The data reported by Sever et al (1977) were similar, athough the conclusions differed. In this study lying and standing noradrenaline levels were measured in 56 hypertensive patients and 44 controls, 9 of whom were hospital staff recruited to improve the age match. The principal result in these age-matched groups were (a) mean plasma noradrenaline levels were very similar in both groups. (b) plasma noradrenaline was related to age in normotensive controls, but (c) in hypertensive patients there was no relationship between plasma noradrenaline and age. The lack of such a relationship was attributed to the presence, among the younger hypertensive patients, of a number of individuals with markedly elevated plasma noradrenaline levels. (Sever et al also stated that a reanalysis of Lake's data gave the same result, namely that an age/plasma noradrenaline relationship was present only in the normotensive subjects, again because a number of young hypertensive patients had disproportionately elevated plasma noradrenaline levels). The observation that high noradrenaline levels tended to be found mainly in younger hypertensive patients suggested that sympathetic overactivity might be important during the early stages of hypertension. Presumably other factors played a more important role in maintaining the raised blood pressure once this had become established. 54 Both these reports included careful descriptions of the age and composition of the control group including the number who were hospital or laboratory staff. However, inspection of table 3.2 suggests that the age of the control subjects had not been regarded as an important factor in the earlier studies. For the practice of recruiting control subjects from young hospital and laboratory staff (see for example, DeQuattro and Champlain 1972) familiar with study procedures tends to reduce stress-induced elevation of plasma noradrenaline (Carruthers, 1970). This would produce low plasma noradrenaline levels in the control groups and give the spurious appearance of elevated noradrenaline levels in the hypertensive patients. This was clearly demonstrated in two studies that used two types of control group, ie (1) composed of medical and laboratory staff, and (2) of non-medical subjects. Plasma noradrenaline levels were lower in the former group (Sever et al, 1977, Jones et al, 1979) . Indeed, Goldstein (1983) has pointed out that studies which use medical or laboratory staff as controls tended to report positive results, while those using medically naive subjects were generally negative. 3.4.6 Other important methodological factors. A number of other factors may influence the outcome of such studies have been cited by Goldstein (1983) in a survey of 78 reported studies in which plasma noradrenaline (or catecholamines) were compared in normotensive and hypertensive subjects. 55 1. The duration of fasting. Fasting is known to increase plasma noradrenaline levels. Studies in which the subjects fasted overnight were significantly more likely to be positive (in this context 'positive' means showing a significant difference between the plasma noradrenaline of the hypertensive and normotensive groups). 2. Duration of recumbency before blood sampling. When patients are supine for longer than 30 minutes, it was more likely that a positive result would be obtained. 3. Prior antihypertensive therapy. In the studies reviewed by Goldstein, the mean time between withdrawal of antihypertensive therapy and plasma sampling for noradrenaline levels was 3 weeks, and ranged from 2 days to 6 weeks. No signiicant difference was found in the likelihood of obtaining a positive result when the patients had been withdrawn from medication for more than 2 weeks compared with less than 2 weeks. However, the frequency of positive results was significantly greater in studies which employed previously treated hypertensives as opposed to previously untreated hypertensives. While it is possible that those already on therapy might simply have had worse hypertension than the untreated ones, it is also possible that the time taken for the effects of prior antihypertensive therapy to wane might be longer than generally supposed. These agents influence sympathetic function both directly and indirectly at both 56 central and peripheral levels, and are often associated with marked changes in plasma catecholamines. Hypertensive patients may show greater increases in catecholamine release in response to provocation compared to normal controls (Roberts et al, 1979; Esler and Nestel, 1973). The stresses of fasting and venepuncture would tend to cause greater increases in blood levels. Hypertensive subjects might also require a longer period of recumbancy for the blood levels to settle. 3.4.7 Evidence for increased sympathetic activity in some young hypertensive patients. In 1977 Sever et al differentiated between the high plasma noradrenaline levels that were associated with increasing age, and the genuinely elevated levels found in some young hypertensive patients. The existence of a young 'hyperadrenergic' group was proposed, and this seemed consistent with haemodynamic studies which appeared to show increased sympathetic activity in younger hypertensives (Lund-Johansen, 1967 and 1980). Esler et al (1977) found that 9 of 21 young hypertensives had elevated plasma noradrenaline levels associated with the haemodynamic features of 'high output' hypertension -ie raised heart rate, cardiac output, and low peripheral resistance. The remainder had normal plasma noradrenaline with the features of 'established' hypertension - high peripheral resistance and normal cardiac output. 57 Subsequently, a large number of studies has been published concerning differences in plasma catecholamine or noradrenaline levels between hypertensive patients and normotensive controls, but only a few have specifically sought evidence for a young hyperadrenergic group. Seventy-eight of these studies were reviewed by Goldstein (1983) who sought evidence of a young, hyperadrenergic subgroup. Of the 78 studies, 64 specifically measured plasma noradrenaline (as opposed to total catecholamines). Overall,plasma noradrenaline was higher in hypertensives than in normotensive groups in 52 of the 64 studies (81%), 25 of which were statistically significant (39%). Analysis of the 51 studies which stated the patients' ages showed that a 'positive' result (ie one showing a signficant difference in plasma noradrenaline levels between hypertensives and normotensives) was much more likely to be found in studies with younger patients, as shown in Table 3.3. Furthermore, of the 27 studies which examined possible correlations between plasma noradrenaline and age, the majority of normotensive groups (15 out of 23, 65%) demonstrated a positive relationship between plasma noradrenaline, whether or not a positive result was obtained. By contrast 16 out of 24 hypertensive groups showed no such correlation (p<0.05; cf Sever et al 1977, section 2.4.5). Although indirect, these results were consistent with the hypothesis that plasma noradrenaline is 58 Table 3.3 The effects of age on the outcome of studies comparing plasma noradrenaline levels in hypertensive and normotensive subjects. group mean No. of No. positive age (years) studies results % <30 5 4 80% 30-39 12 8 67% 40-49 30 7 23% 50+ 4 1 25 {data from Esler et al, 1977; Goldstein et al, 1983) elevated in some young hypertensives. The analysis suggested that these were most likely to be young hypertensives with established hypertension and diastolic pressures consistently above 90mmHg, rather than those with labile or borderline hypertension. Further support was obtained from analysis of plasma noradrenaline data. Goldstein argued that if hyperadrenergic hypertensives were present in variable numbers in the study groups (either by chance or by selection), then the studies in which they were present in decisive numbers would be more likely to have (a) a higher mean plasma noradrenaline level in the hypertensives (b) a greater standard deviation of the plasma noradrenaline level and (c) a positive result. Comparision of 59 the positive and negative studies showed that both the mean plasma noradrenaline levels and the standard deviations were significantly greater in the studies which showed a positive result. Thus the hypothesis that a proportion of young hypertensives have elevated plasma noradrenaline is supported indirectly by the analysis of a large number of studies. 3.4.8 Conclusions. There is no convincing evidence that plasma noradrenaline is generally raised in essential hypertension. Many of the studies which purport to show increased noradrenaline must be doubted for reasons of methodology which include inadequate matching of control groups, the effects of prior antihypertensive therapy and possible differences in noradrenaline handling by hypertensives. There may however be a subgroup of mainly younger hypertensives who exhibit haemodynamic and sometimes somatic features suggestive of sympathetic overactivity, and who have elevated plasma noradrenaline. Thus, in a minority, essential hypertension may be initiated by a primary increase in sympathetic activity, but conclusive evidence is still lacking. Th assumption that plasma noradrenaline reflects sympathetic activity is central to this argument, yet attempts to verify this fundemental relationship experimentally were not made until the late 1970s. However, the growing debate about 60 the role of sympathetic activity in hypertension subsequently focussed a great deal of interest on the nature of sympathetic activity and its effects on plasma noradrenaline. Subsequent work showed that the relationship between them was more complex than previously thought, and led to a more critical appraisal of the validity of plasma noradrenaline as an index of sympathetic activity. This is discussed in the following chapter 61 CHAPTER 4 THE RELATIONSHIP BETWEEN SYMPATHETIC NERVOUS ACTIVITY AND THE PLASMA NORADRENALINE CONCENTRATION 62 4.1 INTRODUCTION The attention focussed on the role of the sympathetic nervous system in essential hypertension was made possible by the availability of improved catecholamine assays. These also provided the means to study sympathetic nervous system function in far greater detail than had hitherto been possible. Prior to this, relatively little was known about the relationship between sympathetic nervous system function and plasma noradrenaline in man, but the interest in hypertension stimulated an enormous research effort which has vastly expanded the understanding of both normal and pathological sympathetic nervous system function. For many years, direct neurographic recordings of sympathetic nerve trunk activity had beer> made in animal studies. In man however, the applicability of this technique is limited essentially to skin and superficial skeletal muscle. Therefore, sympathetic activity is generally estimated indirectly by measuring the plasma concentration of noradrenaline, for this has been assumed to be a valid and reproducible index of sympathetic activity. This assumption has received considerable scrutiny and criticsm in recent years (Folkow et al 1983). The rate of sympathetic nerve firing is clearly a major determinant of plasma noradrenaline, but it is not the only important determinant. It has become inceasingly recognised that the level of noradrenaline in plasma is but the net result 63 of several processes which determine its rate of release and clearance from several vascular beds. The venous effluent of these different beds then mix to form the 'mixed venous plasma noradrenaline'. Although this (and/or the arterial plasma noradrenaline level) theoretically give the best indication of overall sympathetic activity, for practical and ethical reasons it is generally forearm venous blood which is sampled. Here, local factors may sometimes cause plasma noradrenaline to vary independently of changes in the mixed venous pool (Chang et al, 1986) Ideally in order for plasma noradrenaline to be an accurate index of cardiovascular sympathetic activity, the following conditions should be present. 1. The rate of sympathetic nerve firing is the sole determinant of noradrenaline released into the synaptic cleft. 2. The proportion of circulating noradrenaline derived from non-cardiovascular sources is small. 3. The removal of noradrenaline by perisynaptic processes is small; thus the proportion of released noradrenaline which spills over from the synapse to the plasma is large. 4. Di£fusion of noradrenaline from the synapse to the plasma 64 is unimpeded and rapid. 5. There should not be marked regional differences in sympathetic nervous activity in the various vascular beds. In other words, cardiovascular sympathetic activity should be uniform and generalised. 6. Clearance of noradrenaline from the circulation is constant both with and between individuals. This chapter will review the extent to which these different conditions are met. 4.2 SYMPATHETIC NERVE FIRING AND NORADRENALINE RELEASE 4.2.1 Studies in vitro The view that plasma noradrenaline reflects sympathetic activity can be traced back to the results of early in vitro studies. The relationship between sympathetic nerve firing and transmitter release was first studied in single organ preparations, and it was Peart (1949) who first demonstrated the release of noradrenaline into the venous blood of an organ on stimulation of its sympathetic nerve. Later, Brown and Gillespie (1957) showed proportionality between the rate of sympathetic 65 nerve stimulation and transmitter release from the cat spleen. This was also confirmed by Haefley et al in 1965. Similarly renal nerve stimulation provokes a proportionate release of noradrenaline from the kidney (Oliver et al, 1985), and in the pithed rat preparation the increase of plasma noradrenaline varied directly with the frequency and intensity of sympathetic nerve stimulation (Yamaguchi and Kopin, 1979). 4.2.2 Studies in vivo In the anaesthetised open-chested dog stimulation of the cardiac sympathetic nerves caused increases in coronary sinus catecholamine levels which were directly related to the rate of nerve firing (Yamaguchi et al, 1975, 1977). In man direct neurographic recordings from the peroneal nerve have shown that the absolute plasma noradrenaline levels are directly related to the rate of muscle sympathetic nerve activity during supine rest; and both approximately doubled on standing (Delius et al, 1972; Wallin and Sundlof, 1981; Burke et al, 1977) . 4.3 PRESYNAPTIC REGULATION OF NORADRENALINE RELEASE 4.3.1 Presynaptic adrenoceptors There is now good evidence that presynaptic regulation has an important role in modulating the amount of noradrenaline released per impulse from nerve terminals. In the original 66 studies of Brown and Gillespie (1957) the alpha-receptor antagonist phenoxybenzamine increased the stimulation-evoked release of transmitter from the cat spleen, presumably (it is now thought) because of blockade of inhibitory presynaptic alpha-receptors. More recent experimental evidence has been discussed in section 2.4.3. Recent studies in vivo have supported the idea that presynaptic regulation has!a physiological role. Thus, in the anaesthetised dog, the overflow of noradrenaline into coronary sinus increased linearly with increasing frequency of cardiac sympathetic nerve stimulation. The alpha-2 receptor agonist clonidine reduced the stimulation-evoked increases in heart rate, myocardial contractility and noradrenaline release; the greatest effects were noted at lower frequencies of nerve stimulation. Conversely, the alpha receptor antagonist phenoxybenzamine increased both noradrenaline release and cardiac responses with effects predominating at higher stimulation frequencies. These results show that presynaptic modulation of noradrenaline release can significantly affect end organ responses in vivo. (Yamaguchi et al, 1977) Similar studies wre undertaken with the beta-2 receptor agonist isoprenaline and antagonist sotalol. These compounds respectively increased and reduced the stimulation-evoked release of noradrenaline from the dog heart, demonstrating the function of presynaptic facilitatory beta receptors in this species (Yamaguchi et al, 1977). 67 4.3.2 Presynaptic cholinergic receptors in vivo. In a study of presynaptic cholinergic (muscarinic) receptors in dog heart in vivo, Levy and Blattberg (1976) showed that stimulation of the cardiac sympathetic nerves produced the expected increase in myocardial contractility and coronary sinus noradrenaline overflow. However, vagal stimulation reduced noradrenaline overflow by about 30% and this was accompanied by a similar reduction in contractility. These inhibitory effects were abolished by atropine, suggesting that release of sympathetic transmitters could be decisively modulated at the presynaptic level by the parasympathetic nervous system. 4.3.3 Evidence in man. In man, the functional importance of presynaptic receptors in the sympathetic nervous system is still controversial. It has been suggested that adrenaline can modulate noradrenaline release via presynaptic beta-receptors. Despite very low plasma levels, adrenaline can be taken up by the neuronal uptake mechanism and stored in sympathetic nerve endings (Stromblad and Nickerson, 1961). It could then be available for re-release as a 'cotransmitter' along with noradrenaline during sympathetic nerve activity. The adrenaline would be released directly into the synaptic cleft where it could facilitate further noradrenaline release by acting on presynaptic beta-2 receptors (Adler-Grashinsky and Langer, 1975; Majewski et al, 1981). 68 In normal subjects, isoprenaline infusions can raise plasma noradrenaline levels (Vincent et al, 1982); and infusions of adrenaline (producing plasma levels that do not exceed the physiological range) can stimulate noradrenaline release (Brown and Dollery, 1984, Musgrave et al, 1984). Nezu et al (1985) showed persisting increases in both plasma noradrenaline and blood pressure after 30 minute adrenaline infusions which raised heart rate by less than lObpm. The effects on plasma noradrenaline and blood pressure were greater in hypertensives than in normotensive subjects, and could be abolished by propranolol. similar but shorter-lived responses were seen after the release of endogenous adrenaline by the administration of intravenous glucacon. Presynaptic receptors are clearly capable of producing significant effects, but their importance in the everyday functioning of the sympathetic nervous system has not been clarified. 4.4 SPILLOVER AND DIFFUSION OF NEURONALLY-RELEASED NORADRENALINE FROM SYNAPSE TO PLASMA 4.4.1 Uptake and spillover of noradrenaline from the synaptic cleft: studies in vitro. The extent to which neuronally-released noradrenaline 69 passes into plasma is an important determinant of the plasma concentration. Noradrenaline released into the synaptic cleft is inactivated chiefly by neuronal reuptake (uptake 1). Attempts to quatify uptake have been made by blocking the uptake mechanism with cocaine. This results in a roughly four-fold increase in the overflow of transmitter from a variety of organ preparations, suggesting that uptake-1 normally accounts for about 75-80% of released noradrenaline (Langer et al, 1970, Hughes et al, 1972). The amount of released noradrenaline available to spill over is therefore unlikely to be greater than 25%. 4.4.2 Studies in vivo. This figure agrees well with the estimate of 20% spillover in man reported by Hoeldtke et al (1983) who used a kinetic technique. A recent study reported on human volunteers who had been treated with trimethathan and yohimbine (to block autonomic ganglia and vascular alpha receptors, respectively) and then subjected to incremental infusions of noradrenaline to study the relationship between synaptic cleft and plasma concentrations of noradrenaline. In this model, addition of the uptake-1 blocker desiprimine roughly halved the plasma noradrenaline concentration required to elicit a given pressor response. This suggested that uptake-1 accounted for about 50% of the noradrenaline in the vicinity of the receptors (Goldstein et al, 1986), but this proportion would probably have been greater if complete blockade of the uptake mechanism could have been achieved in vivo. On the other hand, it is not clear whether 70 exogenously infused noradenaline (which difuses out of blood vessels to reach synaptic junctions) is equally susceptible to uptake as to noradrenaline which is released locally. These studies suggest that the amount of noradrenaline available to difuse into plasma represents only a small proportion of the total which is released at the nerve terminal. 4.4.3 Diffusion of noradrenaline from synapse to plasma. Once clear of the synapse, the noradrenaline must difuse through medial muscle and intimal layers in order to reach the plasma. Once it has diffused out of the synapse, the amount taken up by extraneuronal uptake into smooth muscle (uptake-2) is probably small (Gillespie, 1973) and indeed may be negligible at normal plasma concentrations (Brown et al, 1982) . Cousineau et al (1980) studied the kinetics of cardiac noradrenaline uptake in closed chest dogs using a multiple tracer dilution technique. This work suggested that the capillary membrane represented an important additional barrier to the outward diffusion of exogenous noradrenaline. If this barrier exists for the diffusion of neuronally released noradrenaline in the opposite direction, the functional consequences could be (a) retention of transmitter in the extracellular space (b) potentiation of presynaptic effects (c) a further concentration gradient between synaptic and plasma noradrenaline. However, these findings have not been duplicated by others and their significance is uncertain. Furthermore, 71 there is no information about diffusion in non-cardiac tissue. 4.5 THE EFFECTS OF INCREASED SYMPATHETIC ACTVITY ON PLASMA NORADRNALINE LEVELS 4.5.1 Introduction Although the relatively small proportion of neuronally released noradrenaline which diffuses out of the synaptic cleft suggests that changes in sympathetic activity will be reflected in plasma in dampened form, there is now a considerable body of evidence that plasma noradrenaline levels vary with changes in sympathetic activity, at least where these changes are relatively large. Noradrenaline in the body is distributed in the sympathetic nerve endings, brain and adrenal medulla. The contribution of cerebral noradrenaline to circulating levels is insignificant since hardly any reaches the blood unmetabolised (Glowinski et al, 1965). The adrenal medula secretes noradrenaline into the blood stream, but its contribution to circulating noradrenaline is very small, probably between 2% and 7.5% (Brown et al, 1981; Planz et al, 1978) and adrenalectomy has no significant effect on noradrenaline levels. Thus at rest approximately 95% of circulating noradrenaline originates from sympathetic nerve endings. 72 In general, experimental assessment of sympathetic activity man in requires the measurement of plasma noradrenaline and adrenaline during maneouvres believed to be associated with changes in sympathetic activity. The stimuli often consist of circulatory, psychological or environmental stresses. They are useful in provoking relatively large changes in sympathetic activity, may be poorly suited to studying quantitative aspects of sympathetic physiology since many are relatively gross stimuli that provoke an all-or-none response. 4.5.2 Experimental sympathetic stimuli that raise plasma noradrenaline: general. Noradrenaline release can be provoked experimentally by a number of different stimuli. In general the increase of plasma noradrenaline is related to the intensity of the activity. Thus, Lake et al (1976) noted progressively higher plasma noradrenaline levels as patients changed from supine rest (0.29 ng/ml) to sitting (0.46 ng/ml) to standing (0.54 ng/ml) and finally to isometric effort while upright (0.78 ng/ml). Similar results were reported by Reid et al (1978) who also noted a significant linear relationship between plasma noradrenaline and systolic blood pressure, and plasma noradrenaline and heart rate. This relationship appeared true for individuals as well as for the whole group. These results are shown in Table 4.1. 4.5.3 Syirpathetic stimuli: exercise. 73 Dynamic exercise is a well documented stimulus for increasing plasma noradrenaline levels. The magnitude of Table 4.1 Plasma noradrenaline levels during various activities in 8 hypertensive subjects (mean + sd). activity______plasma NA (ng/ml) sleeping 0..27 + 0..03 supine rest 0..51 + 0..07 sitting 0..56 + 0..05 standing 0..68 + 0..05 walking 0..79 + 0..08 bicycling 2,.32 + 0..47 (data from Reid et al, 1978) the rise of plasma noradrenaline is related to the intensity of exercise or workload rather than the type of exercise performed. Heavy exercise may raise plasma noradrenaline levels by ten-fold or more (see table 4.1). By contrast, isometric exercise raises blood pressure but has a lesser effect on the heart rate. The increment of plasma noradrenaline is usually small, typically 20-50% (Kozlowski et al, 1973; Watson, Hamilton et al, 1979; Watson, Page et al, 1979; Robertson et al, 1979; Floras et al, 1986), but is sometimes greater if the exercise is prolonged. 74 4.5.4 Orthostasis. Standing up abruptly reduces the venous return to the heart, causing activation of sympathetic reflexes. These include the baroreflex which is mediated by carotid sinus and aortic arch receptors, and the low-pressure cardiopulmonary 'volume' receptors. They provide a tonically inhibitory input to the medullary vasomotor centre, whose outflow increases venous tone and peripheral arteriolar resistence to counteract fall in blood pressure. Blood pressure remains the same or may rise slightly, heart rate increases and plasma noradrenaline increases by about 50 - 100%. Selective stimulation of subgroups of baroreceptors has shown differential patterns of sympathetic response. Hancia et al (1979) used a variable pressure neck chamber to produce changes in the degree of carotid sinus baroreceptor stimulation. Selective reduction of carotid sinus stimulation produced activation of the baroreflex leading to marked increases in arterial pressure and heart rate but no increase in plasma noradrenline. By contrast, head-up tilt greatly raised the plasma noradrenaline but did not increase the heart rate or the arterial pressure. The effects of head-up tilt can be closely reproduced by selective reduction of cardiopulmonary (volume) receptor activation in a lower body negative pressure chamber. There is little effect on blood pressure and heart rate -presumably because the purpose of the reflex is to maintain cardiac filling presure at its previous level; the marked 75 increase in plasma noradrenaline reflects constriction of venous capacitance beds. (Sundlof and Wallin, 1978; Hjemdahl et al, 1982). These differential responses are caused by variations in regional sympathetic outflow. 4.5.5 Mental stress. Mental stress can be a potent stimulus of sympathetic activity as shown by tachycardia and a raised blood pressure, but a corresponding rise in plasma noradrenaline is not consistently seen, particularly with mental arithmatic (Floras et al, 1986, Le Blanc et al 1979). Other maneouvres produce varying results: Stroop's colour-word conflict test failed to raise noradrenaline or adrenaline in hypertensive and normotensive subjects despite convincing changes in heart rate and blood pressure (Hjemdahl and Eliasson 1979) . However, mental arithmatic performed to the accomaniment of prolonged, loud and unpleasant noise for 30 minutes raised both urinary and plasma catecholamines and their metabolites (Januszewisz et al, 1979). The anticipation of pain, such as venepunction (Lake et al, 1976) and dental extractions (Goldstein et al, 1982) is a more potent stimulus for catecholamine release. Again, it is likely that noradrenaline release is related to the intensity of the stimulus. 4.5.6 Environmental stress. Exposure to low temperature, and manoeuvres such as the 76 cold pressor test appear to cause sympathetic activation; although in the latter case, pain and its associated psychological stress may also be important. Exposure to cold by emersion in cold water (Zenner et al, 1980) or in cold room air (Epstein et al, 1969) raises blood pressure by increasing peripheral resistence, whereas heart rate increases only slightly (Hattenhauer and Neil, 1975). Plasma noradrenaline may rise up to three-fold (Johnson et al, 1977). Sympathetic activation probably accounts for both the haemodynamic changes and the rise in noradrenaline, but whether some of this noradrenaline is released as a consequence of muscle shivering as well as arteriolar constriction is not clear. 4.5.7 Vasodilator drugs. Vasodilator drugs may cause activation of baroreceptor-mediated reflexes leading to rapid and marked increases in plasma noradrenaline. Such effects have been reported for nitroprusside, prazosin, phentolamine, nifedipine and hydralazine (Reid, 1979; Lin et al 1978; Murphy et al 1983). 4.5.8 Myocardial infarction. Myocardial infarction causes marked elevation of plasma catecholamines in animals and in man. In anaesthetised dogs with coronary artery ligation, the magnitude of the rise in plasma catecholamines was related to the size of the experimentally induced infarct (Karlsberg et al, 1979) In man, both 77 catecholamines can rise to enormously high values, but adrenaline levels may be up to ten-fold higher than noradrenaline (Christensen and Videback, 1974; Muller and Ayres, 1980; Worstman et al, 1984) There is indirect evidence in man to suggest that the magnitude of the rise in catecholamines is related to the size of the infarct (Little et al, 1985), as shown in dogs by Karlsberg (1979). 4.5.9 Miscellaneous Various other stimuli have been shown to raise plasma noradrenaline, presumably by increasing release. These include caffeine (Robertson et al, 1979) cigarette smoking (Cryer et al, 1976), diuretic therapy (Esler 1982), a low sodium diet (Robertson et al, 1979) and hypoglycaemia (Garber et al, 1982). There is also an important diurnal variation in plasma levels (Saar and Gordon, 1979). 4.6 THE EFFECTS OF REDUCED SYMPATHETIC ACTIVITY ON PLASMA NORADRENALINE LEVELS 4.6.1 Animal evidence In laboratory animals, the release of catecholamines by kidney and spleen ceases when the sympathetic nerves to these organs are cut (Brown et al, 1961; Stephenson et al, 1982). 78 Similarly/ physical disruption of central sympathetic outflow by pithing or by cervical spinal cord transection/ causes plasma noradrenaline levels to fall (Yamaguchi and Kopin 1979; Tibbs et al/ 1979) . Pharmacological means of reducing sympathetic outflow, either by ganglion blockade (Reid and Kopin, 1977), or by inhibiting the release of noradrenaline from sympathetic nerve endings (Kvetnansky et al, 1979) also causes a reduction in plasma noradrenaline. 4.6.2 Evidence in man. In man, circulating noradrenaline levels are reduced by a number of drugs which attenuate sympathetic activity at various levels. These include clonidine which inhibits central sympathetic outflow (Schmitt and Schmitt, 1969; Reid et al, 1976); the ganglion blocker pentolinium (Louis et al, 1973; Murphy et al, 1983); and debrisoquine, which interferes with the post-ganglionic trasmission of sympathetic activity (Flammer et al, 1979). Patients who have undergone unilateral sympathectomy for Raynaud's phejomenon have a low plasma noradrenaline in the venous effluent of the sympathectomised arm (Nielson et al, 1980), and patients rendered tetraplegic by traumatic cervical spinal cord section have a low circulating noradrenaline (Reid et al, 1976). 4.6.3 Progressive autonomic failure. This is a neurological disorder which may occur in 79 isolation (idiopathic orthostatic hypotension), as a complication of diabetes mellitus, or as part of a more extensive central neurological disorder such as the Shy-Drager syndrome. In such cases a severe postural fall in blood pressure takes place when the patient attempts to stand, and the usual rise of plasma noradrenaline is absent or blunted. In those with central neurological syndromes, resting plasma noradrenaline is usually normal since the peripheral sympathetic nervous system is intact but not appropriately activated by postural stress. By contrast, patients with idiopathic othostatic hypotension have subnormal resting plasma noradrenaline levels which indicate actual degeneration of the peripheral sympathetic nervous system (Polinski et al, 1976; Cryer, 1980). 4.7 CLEARANCE OF NORADRENALINE FROM THE PLASMA 4.7.1 Introduction. The estimation of sympathetic nervous activity by measurement of the noradrenaline concentration in plasma assumes a constant relationship between this, the synaptic cleft noradrenaline concentration, and the rate of neuronal noradrenaline release. However, the plasma noradrenaline level is determined by the rate of its removal from the circulation by metabolic clearance as well as by the rate of release. Clearance is achieved by the combined processes of neuronal uptake into sympathetic nerves (uptake-1), extraneuronal uptake by other 80 tissues such as vascular smooth muscle and endothelium (uptake-2), and metabolic conversion by O-methylation, oxidative deamination and conjugation (Kopin, 1979). Studies in animals and in man with inhibitors of neuronal uptake point to the predominance of this mechanism, and to a lesser extent of O-methylation, as the principal determinants of the metabolic clearance rate. Extraneuronal uptake seems unimportant at normal plasma concentrations (Hertting et al, 1961; Esler et al, 1981a; Brown et al, 1982). 4.7.2 Methodology A number of kinetic methods have been developed to measure the noradrenaline metabolic clearance rate, and to derive from this data the rate of input of noradrenaline into the circulation. The latter is presumably a more accurate reflection of neuronal release. Measurement of noradrenaline clearance generally depends upon infusions of 1-noradrenaline to steady-state, and methods employing both the labelled and unlabelled 1-isomer have been described. Use of the racemic mixture may lead to errors because the uptake-1 mechanism has a lower affinity for the d-isomer (Esler, 1982). A problem common to all methods is that the use of too high an infusion rate may alter haemodynamics and cause reflex changes in sympathetic activity which may produce erroneous measurements. Thus low infusion rates of labelled noradrenaline having high specific activity are preferable. 81 4.7.3 Published studies. Table 4.2 shows the results obtained for noradrenaline plasma clearance in normal subjects obtained by different investigators. Table 4.2 Noradrenaline plasma half-life and metabolic clearance rates in published studies. clearance halflife author/year method 1-NA (L/min/m2) (mins) Silverberg 1978 infusion unlabelled 1.6 2.4 Fitzgerald 1981 infusion unlabelled 2.1 2.1-2.4 Rubin 1982 infusion unlabelled 2.5 - Brown 1982 infusion unlabelled 2.8 - Grimm 1980 infusion unlabelled 2.8 1.8 Ghione 1978 bolus 3H-labelled 1.1 - Esler 1979 infusion 3H-labelled 1.3 2.0 Esler 1981 infusion 3H-labelled 1.2-1.5 - Clearance rates obtained with infusions of unlabelled noradrenaline are consistently two-fold greater than the results obtained with radiotracer techniques. These discrepancies probably arise from methodological differences in assaying plasma noradrenaline, suppression of endogenous noradrenaline release, and possibly noradrenaline-induced changes in regional 82 haemodynamics. Among normal and hypertensive subjects, the range of noradrenaline clearance rates is quite wide with three or four-fold differences between the lower and upper ends of the range. Clearance rates tend to fall by up to 40% with increase in age (Esler, Skews et al, 1981; Rubin et al, 1982) . 4.7.4 Effects of drugs. The variability in clearance rates may be further increased by the effects of various drugs. The tricyclic antidepressant desipramine reduced plasma noradrenaline release, but since clearance was also reduced, the net effect was that plasma noradrenaline levels remained unchanged. Beta blockers such as propranolol and oxyprenolol increased plasma noradrenaline levels secondary to a reduction in noradrenaline clearance; yet noradrenaline release remained unchanged. Conversely, in the case of clonidine, noradrenaline clearance was unchanged, thus the fall in plasma noradrenaline levels genuinely reflected a reduction in noradrenaline release. Similar results were obtained for frusemide (Esler, 1982). Drugs which affect cardiac output may also affect noradrenaline clearance by altering the perfusion of organs and tissues which extract noradrenaline from plasma. Thus, both adrenaline (Cryer et al, 1980) and isoprenaline (Brown et al, 1982) increase catecholamine clearance. This may also occur 83 with other vasodilator drugs. Conversely cardiodepressant drugs such as propranolol and oxypranolol reduce noradrenaline clearance (Cryer et al, 1980; Esler, 1982). 4.7.5 Conclusions. These kinetic data emphasise the limitations of static plasma noradrenaline levels as a means of assessing sympathetic nervous activity. In particular, the degree of variability of noradrenaline plasma clearance rate, a well as other factors such as the volume of distrubution (Fitzgerald et al, 1979) raise doubts about the usefulness of plasma noradrenaline levels for comparing sympathetic nervous activity between individuals. 84 CHAPTER 5 GENERAL METHODOLOGY 85 5.1 INTRODUCTION This chapter describes the general methodology used in the studies which are reported in chapters 6 to 9 inclusive. These studies can be regarded as non-invasive since intravascular cannulation was limited to one or two peripheral venous catheters required for blood sampling and drug infusion. The techniques used for haemodynamic measurements, blood sampling, catecholamine assay, drug dosage and administration, cold stress, ethical aspects and the general conduct of the study will be considered. (Details of the invasive studies involving cardiac catheterisation and cardiac pacing will be described in chapters 11 and 12). 5.2 SUBJECTS A total of 14 volunteer subjects took part in the non-invasive studies. All were male and were medical or scientific staff of the Department of Clinical Pharmacology of the Royal Postgraduate Medical School. The ages ranged from 22 to 44. All were in good health as judged by clinical history, physical examination and a normal resting electrocardiogram. None had ever suffered from any important clinical disorder, and specifically, all subjects were free from asthma, hypertension and rhythm disturbance. Two subjects suffered from seasonal hay 86 fever, consisting of allergic rhinitis and conjunctivitis. Two other subjects were regular cigarette smokers but abstained for at least 12 hours before studies. Written informed consent was obtained from all subjects after a full verbal and written explanation had been given about the purpose, methods and possible risks of the studies. Experimental protocols were approved by the Research Ethics Committee of the Royal Postgraduate Medical School. 5.3 GENERAL CONDUCT OF THE STUDIES All subjects abstained from tobacco, alcohol and caffine for at least 12 hours before each study. With the exception of the cold stress study, all procedures took place in a purpose-designed clinical laboratory maintained at 20 - 22 degrees Centigrade (C). The studies began at 0800 to 0900, approximately 2 hours after a light breakfast (toast or cereal) in order to avoid hypoglycaemia which might lead to adrenaline release (Gerber et al, 1976). 5.4 DRUG ADMINISTRATION 5.4.1 Intravenous infusions All infusions were given through a short 21 guage cannula 87 (Venflon or Medicut) placed in a large antecubital vein following local infiltration of 1% lignocaine in the arm contralateral to that used for blood pressure measurements. The Infusions were controlled by a variable-rate, electrically powered slow-infusion pump driving a 30ml or 50ml syringe. Drug infusion rates could therefore be conveniently changed simply by altering the rate setting. 5.4.2 Noradrenaline infusions L-noradrenaline (Levophed, Winthrop) was diluted with 0.9% sodium chloride for intravenous infusion. The infusion rates were 0.01, 0.02, 0.03, 0.05, 0.07, 0.lOug/kg/minute. The duration of the infusion was ten minutes at each infusion rate. 5.4.3 Tyramine infusions Tyramine hydrochloride (Sigma) was obtained commercially, sterilised and packed in vials suitable for intravenous administration by the pharmacy of Hammersmith Hospital. Each vial contained 5mg of tyramine HC1 base in 5ml of sterile water. A variety of infusion rates were used from 2.5 to 25 um/kg/minute and the durations of each infusion was between 10 and 30 minutes. Further details will be given with the individual studies. 5.4.4 Oral propranolol 88 Oral propranolol (Inderal, ICI) was administered as a single oral dose of four 40mg tablets. 5.5 COLD STRESS Cold stress studies began at 0900 and took place in a 3m x 3m refrigerated laboratory cold room maintained at 4 C. The subjects lay on a matress placed on the floor and were covered with blankets. A venous sampling cannula was inserted in an antecubital vein, and the sphygmomanometer cuff placed on the contralateral arm. After 40 minutes of supine rest, baseline blood pressure, heart rate and blood samples (for noradrenaline and adrenaline) were taken twice within five minutes. The subjects, who were dressed only in shorts were then abruptly exposed to the cold air temperature by removal of the blankets. Exposure to cold air continued for 30 minutes; further blood samples and haemodynamic measurements were taken at 5, 10, 15, 20, 25, and 30 minutes. The subjects were closely observed throughout. There was no overt shivering although superficial fasciculation was observed in some subjects. 89 5.6 HAEMODYNAMIC MEASUREMENTS 5.6.1.Blood pressure. In the non-invasive studies, brachial blood pressure was measured using a semi-automatic sphygmomanometer, a Roche Arteriosonde 1217, taking the mean of three successive determinations. This could generally be completed within one minute. The accuracy of the Arteriosonde was checked by comparing it with simultaneously taken blood pressure determinations using a Hawksley random-zero sphygmomanometer connected to the same arm cuff using a Y piece attachment. Systolic and diastolic blood pressure recordings (mean of three readings) in 12 hypertensive patients and 8 normal subjects are shown in figures 5-1 and 5-2. The results obtained using the two sphygmomanometers were very similar. Seventeen of twenty systolic readings and, eighteen of twenty diastolic readings were within 3 mmHg; the correllation coefficient r was 0.98 for diastolic, and 1.0 for systolic pressures. 5.6.2 Heart rate. The ECG was recorded continuously on a Grass Polygraph 7D and the heart rate was measured at relevent data points by 90 systolic BP (m m Hg) Hawkesley Figure 5-1. Systolic blood pressure readings from 12 hypertensive patients and 8 normotensive subjects, obtained simoultaneously on Hawkesley and Arteriosonde sphygmomanometers. Each point is the mean of three readings. 91 Figure 5-2. Diastolic blood pressure readings from 12 hypertensive patients and 8 normotensive subjects, obtained simoultaneously on Hawkesley and Arteriosonde sphygmomanometers. Each point is the mean of three readings. 92 counting directly from the paper record for 30 seconds. 5.6.3 Systolic time intervals. Systolic time intervals were measured from simultaneous recordings of the electrocardiogram, the phonocardiogram and the indirect external carotid pulse (Lewis et al, 1977). The following intervals were derived:- (a) Total electromechanical systole (QS2), taken from the earliest deflection of the Q wave in standard lead II of the resting electrocardiogram to the first high-frequency deflections of the second heart sound (assumed to be aortic valve closure) on the phonocardiogram. (b) Left ventricular ejection time (LVET). This was measured from the onset of the rapid upstroke of the carotid pulse wave to its incisural notch. (c) Pre-ejection period (PEP). This is the interval between the onset of electrical events and the onset of left ventricular (LV) contraction, marked by the inscription of the electrocardiographic Q wave and the upstroke of the carotid pulse, respectively. However, the PEP cannot be measured directly from the paper trace because the onset of the carotid 93 pulse wave is delayed (with respect to the ECG) due to the time taken for transmission of the LV pressure wave to the neck, usually 20-40 ms. PEP is therefore derived by subtracting LVET from QS2. The right external carotid pulse was recorded at the point of maximum pulsation using a Cambridge transducer air-coupled via rigid polythene tubing of 3 mm internal diameter to a funnel whose open end was 2 cm in diameter. A Cambridge Ekoline phonocardiogram was positioned on the chest wall to give optimum definition of aortic valve closure, usually around the 2nd left or right intercostal space. All recordings were made on Cambridge Ekophonocardiographic equipment at a paper speed of 100 mm/sec. The intervals from ten beats (consecutive if possible) were measured to the nearest 5ms and the results averaged. The results were corrected for heart rate using the regression equations proposed by Weissler et al (1968), and expressed as the index ie, PEPI, LVETI, QS2I. The regression equations of Weissler et al are shown in table 5.1. 94 TABLE 5.1 Calculation of systolic time interval index values from regression equations. STI sex equation normal value QS2I: M QS2l=QS2+2.1HR 546+14 F QS2l=QS2+2.OHR 549+14 LVETI: M LVETI=LVET+1.7HR 413+10 F LVETI=LVET+1.6HR 418+11 PEPI: M PEPI=PEP+0.4HR 131+10 F PEPI=PEP+0.4HR 133+10 abbreviations: M,male; F,female; HR,heart rate. 5.7 PLASMA SAMPLING 5.7.1 Blood samples A 19 guage butterfly cannula was placed in a large superficial forearm vein using 1% Lignocaine local anaesthesia. Patency was maintained by flushing with heparinised saline (100 units per litre) after blood sampling had taken place, and at intervals of 5 minutes otherwise. 5.7.2 Handling and storage 95 Blood samples of 2 ml were drawn into chilled heparinised syringes which were immediately placed in an ice/water mix until centrifuged, usually within 10 or 15 minutes. Centrifugation was performed at 3,000rpm for 5 minutes, in a cold centrifuge maintained at 4 C. The chilled plasma was carefully pipetted into 1ml storage tubes. Great care was taken to avoid haemolysis of the samples of the inclusion of erythrocytes since haemoglobin has been shown to inhibit the enzyme catechol-o-methyltransferase used in the catecholamine assay. These samples were stored at -20 C for the remainder of the experiment, and then transferred for storage at -80 C until assayed. 96 5.8 MEASUREMENT OF PLASMA CATECHOLAMINES 5.8.1 Introduction At present, the two methods of choice for assaying plasma catecholamines are radioenzymatic assays, and high pressure liquid chromatography (HPLC). The earlier fluorimetric methods are no longer widely used. In the present studies, the assay was a double-isotope radioenzymatic technique using the enzyme catechol-o-methyltransferase (COMT). 5.8.2 Radioenzymatic assays: general points. These assays are based on conversion of the substance being measured to a radiolabelled derivative. This occurs in the presence of a specific enzyme and a labelled co-substrate. If only one labelled co-substrate is used, the technique is descibed as a single-isotope assay. These assays are more limited in their potential application than radioimmunoassays since they depend on the existence of a suitable enzyme. Many radioenzymatic assays use a methyltransferase and a labelled methyl group (CH3) donor in the form of S-adenosyl-l-methionine (SAM). The reaction is: X + *SAM = *CH3X + S-adenosyl-L-homocystine ...where X is the unknown compound, and the asterisk (*) 97 indicates the presence of the radiolabelled group. Unlike radioimmunoassays where the amount of label used is deliberately limited so that the unknown competes with it for the antibody, in radioenzymatic assays the amount of label must be saturating so that the reaction rate depends solely on the concentration of the unknown. The amount of labelled product formed therefore varies in direct relation to the amount of unknown substrate present. It is consequently necessary, after termination of the enzyme reaction, to separate the now radiolabelled substrate from the large excess of label. Furthermore, in the case of assays which measure the individual catecholamines adrenaline and noradrenaline, their methylated products (metanephrine and normetanephrine) also need to be separated from each other. This involves several steps from which recovery is relatively low (approximately 50%) and may be variable. This must be corrected for in order to achieve adequate precision. In the assay described below, methanephrine and normetanephrine are separated using thin layer chromatography on silica gel; this is not a feature of all published catecholamine assays. However, as well as reducing the background counts due to the remaining excess label in the reaction mix, it also improves specificity by eliminating any other (labelled) methylated compounds generated by contaminating nonspecific methyltransferase enzymes in the COMT preparation. 98 Apart from the COMT assay, the other major radioenzymatic method for assaying noradrenaline is based on phenylethanolamine N-methyltransferase. This has the disadvantage that it only assays noradrenaline, and, in its original form, was less sensitive and precise than the COMT assay. 5.8.3 Principles of the double-isotope radioenzymatic assay. In double-isotope techniques, two different radiolabels are used. Generally, a known, tracer amount of 3H-labelled X is added to the unknown X at the start of the assay. This is so that any losses of the 3H-tracer which occur during the assay can be used to corect for losses of the unknown. Both the unknown and known (ie labelled) X are then converted to 14C-methyl-X by a methyltransferase, using 14C-SAM as the labelled co-substrate. However, since 14C-isotopes are up to 1000-fold less 'hot' than 3H-isotopes, these assays are often far less sensitive than single-isotope assays using 3H-SAM alone. (The alternative of using a 14C-labelled tracer is not appropriate, for 14C-isotopes are so 'cold' that the amount of tracer required to yield a measurable number of counts would be much higher than the amount of the unknown. In this case, most of the 3H-counts would derive from the methylated tracer and not the unknown). 99 5.9 PRINCIPAL STEPS IN THE PRESENT ASSAY 5.9.1 Introduction. The present assay was developed by Brown and Jenner in the Department of Clinical Pharmacology of the Royal Postgraduate Medical School, and reported in 1981. The same assay was also used in the Department of Clinical Pharmacology at the Centre Hospitallier Universitaire Henri-Mondor, University of Paris. 5.9.2 Specific modifications. of . . In the present COMT assay the problem^low sensitivty due to the 'coldness' of the 14C isotope was overcome by using 14C-methanephrines as the tracer, added after termination of the methylation reaction. In this way, none of the 3H counts could be derived from the tracer, however great its concentration. Therefore, much higher concentrations of the 14C-labelled tracer could be used so that the sensitivity of the assay would not be compromised. As 14C-metanephrines are not commercially available, it was therefore necessary to form them in a separate tube as the initial step of the assay. Separate aliquots of each plasma sample to be assayed were incubated in the presence of either (a) COMT + 3H-SAM, in duplicate, or (b) COMT + 14C-SAM plus standard concentrations of 'cold' noradrenaline and 100 adrenaline. The latter were converted to standards of 14C-normetanephrine and 14C-metanephrine and added back to the 3H tubes after the methylation. (Excess 'cold' SAM is added to the 14C tubes before this transfer in order to safeguard any- remaining catecholamine standards from methylation by 3H-SAM after transfer). This method of addition of 14C-labelled catecholamines to the 3H-labelled unknown samples has several advantages. These are: 1. The plasma concentrations of catecholamines are calculated from the ratio of 3H counts to 14C counts. Recovery of the radiolabelled isotopes from the various steps of procedure may be variable, but will be the same for both isotopes. Losses are corrected for since the ratio of 3H counts (unknown) to 14C counts (standards) will remain the same. 2. Each sample carries its own 14C-labelled internal standard. Any marked variation from the average level of 14C count in individual tubes may indicate incorrect assaying or other problems, and may be helpful in deciding whether that result should be discarded. 3. Plasma tends to inhibit methylation by COMT. The efficiency of this reaction is only approximately 50% and may vary widely among different plasma samples. However, the use of 14C labelled standards corrects for the efficiency of formation of the 3H-metanepharines from the unknown samples. 101 5.9.3 Sequential steps: initial incubation mix. The first step in the assay sequence involves preparation of the incubation 'mix' which is to be added to the unknown plasma samples. This contains: (a) The enzyme catechol o-methyltransferase, prepared from rat liver (Axelrod and Tomchick, 1958). (b) Magnesium chloride. Mg++ ions are an absolute reqirement for COMT activity. (c) The ingredients in the incubation mix (other than COMT and SAM) are magnesium chloride (MgC12 ). Mg++ ions are an absolute requirement for COMT activity. (d) EGTA - This chelates calcium ions (which inhibits COMT activity) but not Mg ++ ions unlike EDTA which chelates). (e) Benzylhydroxylamine - this inhibits the enzyme DOPA decarboxylase which frequently contaminates the COMT preparation. In plasma, DOPA is present in much higher concentrations than catecholamines; and if decarboxylated to dopamine the latter is O-methylated by the COMT. This can interfere with adrenaline or isoprenaline measurements. 102 (f) Tris buffer - this buffers the reaction at the enzymes maximum pH, 8.4. At this point, the incubation mix is divided into two parts. To one part is added 3H-SAM as the co-substrate. To the other part is added 14C-SAM plus standards of noradrenaline and adrenaline to generate the 14C-labelled normetanephrine and metanephrine tracers. 5.9.4 Methylation reaction. The 3H and 14C incubation mixes are added to the unknown plasma samples and incubated at 25 C for 1.5 hours for methylation of catecholamines to the respective metanephrines. At the end of this time the tubes are cooled by transfer to an ice-bath, and the reaction is terminated in the 14C samples by the addition of a great excess of 'cold' SAM which prevents any further formation of 14C-labelled metanepharine. Aliquots from the 14C tubes are then added to the 3H tubes, which now contain (a) 3H-labelled unknown metanephrines, and (b) 14C-labelled metanephrine standards. 5.9.5 Extraction of metanephrines. At this point, the labelled metanepharines are in aqueous solution in the incubation mix which contains large amounts of 103 labelled SAM together with 'cold' SAM and the other constituents noted above. The metanephrines are extracted from this solution by the addition of tetraphenylboron (TPB) which forms an ion-pair complex with the metanephrines, rendering them soluble in organic solvents. The TPB-metanepharine complex is then extracted with diethylether. This complex is unstable at acid pH (catecholamines and their methylated derivatives are weak alkalis with a pK of 9-10 and are fully ionised at physiological pH and below), so that that the TPB plus metanepharine complex can be back-extracted into a much smaller volume of acid. The second extraction also helps to eliminate residual excess 3H-SAM and also concentrates the aqueous phase to a small volume suitable for thin layer silica gel chromatography. 5.9.6 Separation of catecholamines This is achieved using thin-layer silica gel chromatography. The metanephrines, in solution in a small volume of acid, are spotted onto the chomatography plates and allowed to evaporate to dryness. When placed in the solvent the metanephrines migrate at different rates. These form discrete bands which can be visualised under ultraviolet light and scraped off into different tubes as required for the final stage of the assay. The leading band is 3-methoxytyramine (dopamine) followed by metanephrine (adrenaline) followed by normetanephrine (noradrenaline). 104 The metanephrines are oxidised by periodate to replace the terminal CH3 or NH2 group with a carboxyl group (COOH) to form vanilic acid. This has an acid pK so that the addition of glacial acedic acid allows it to be extracted out of the aqueous phase into toluene. Any surviving 3H-SAM or other radiolabelled contaminants are not similarly altered by the periodate and remain in the aqueous phase. 5.9.7 Counting. The toluene, containing the 14C and 3H-labelled catecholamine derivatives (now in the form of vanillins) is added to 'Permafluor' scintillation mix for counting. 5.9.9 Calculation of the unknown catecholamine concentration. The 3H/14C ratio for each sample is compared to the ratio for samples for which a known amount (usually 1.Ong) of noradrenaline or adrenaline was added to the 14C-tubes prior to the incubation. The unknown catecholamine concentration is given by. Unknown CA concentration = ratio 1 - ratio 2 ------x xo ratio 3 - ratio 1 105 Where ratio 1 is the 3H/14C ratio for the sample. ratio 2 is the 3H/14C ratio for the blank tubes, ratio 3 is the 3H/14C ratio for the standard The intra-assay coefficient of variation was 3-5%, and the interassay coefficient of variation was 6-8% over the period that these studies took place. A copy of the assay instruction sheet used in the Department of Clinical Pharmacology of the Royal Postgraduate Medical School is included in the Appendix. All the catecholamine results presented in chapters 6 to 9 inclusive were assayed by the author; those in chapters 11 and 12 were assayed by technical staff of the Department de Pharmacologie Clinique in the Centre Hospitallier Universitaire Henri-Mondor, Paris. 106 CHAPTER 6 THE RELATIONSHIP BETWEEN BLOOD PRESSURE AND PLASMA NORADRENALINE LEVELS AFTER RELEASE OF ENDOGENOUS NEURONAL NORADRENALINE BY TYRAMINE 107 6.1 INTRODUCTION The sensitivity of plasma noradrenaline levels to small changes in sympathetic activity is not known, nor is it clearly established that there is any consistent relationship between given changes in sympathetic activity and the consequent changes in the plasma noradrenaline. A number of reports have drawn attention to increases in plasma noradrenaline produced by various haemodynamic, physical and psychological stimuli. These have shown that the stimulation of certain sympathetic reflexes leads to noradrenaline release, and that the increment in plasma noradrenaline is in general related to the intensity of the stimulus. Relatively large changes in sympathetic activity can be reliably detected, but the ability to detect small changes depends on a number of factors. Firstly, it is not clear whether the spillover of noradrenaline from the synaptic cleft and its subsequent diffusion into the bloodstream increases in proportion to the rate of noradrenaline release from sympathetic neurones. Up to 80% of noradrenaline released into synaptic clefts is removed by neuronal reuptake; about 20% spills over into plasma, but it is not known if these' proportions remain constant whatever the rate of nerve firing. The neuronal uptake mechanism is saturable; it is possible that uptake mechanisms become overloaded at high rates of nerve firng, with a consequent increase in spillover. If this were the case, the relationship between neuronal release and plasma levels woud be unpredictable; plasma noradrenaline 108 levels would not be quantitatively related to neuronal release even within an individual. Another factor which may complicate the detection of small changes in sympathetic activity is the variabilty of plasma noradrenaline levels. This is considerable, both within and between individuals, and results in standard deviations which are up to 50% of the mean group value. It is caused by biological variation and by the poor accuracy and precision of some assays. In this context the reliable detection of small changes in plasma noradrenaline requires careful control of the external variables, as well as accuracy, precision and sensitivity in the catecholamine assay. As noted elsewhere, conventional psychological and physiological sympathetic stimuli are not well suited to producing small incremental changes in sympathetic activity since they are largely uall-or-none" stimuli. In addition, many involve changes in posture and central haemodynamics, or release adrenaline as well as noradrenaline, and this may complicate the interpretation of haemodynamic data. In the present experiment, these difficulties were avoided by the use of the indirectly-acting sympathomimetic amine tyramine, which is structurally similar to noradrenaline (figure 6-1). This was given as a continuous infusion to produce a controlled and graded pressor effect. The mechanism of action of tyramine is not fully understood (Rapoport et al, 1981) but it 109 is agreed that tyramine is taken up into sympathetic nerve endings by the neuronal reuptake mechanism and actively incorporated into noradrenaline storage vesicles where it displaces noradrenaline into the cytosol. This displaced noradrenaline diffuses out of the sympathetic nerve terminal and into the synaptic cleft where it is available to bind postsynaptic receptors, act as a substrate for neuronal reuptake, or spill over into the plasma. For experimental purposes, therefore, the effects of tyramine were regarded as being comparable to endogenous noradrenaline release. The infusion rates for tyramine were chosen to span the range from just below the threshold for pressor effects to a dose that produced an obvious effect. In this way it was possible to relate a series of changes in blood pressure to the corresponding changes in plasma noradrenaline. The effects of tyramine were also compared with those of infused 1-noradrenaline to determine if there were differences in the plasma noradrenaline/blood pressure relationship resulting from exogenous as opposed to endogenous noradrenaline delivery. A subsidiary aim of this study was to assess whether such a comparison could give an estimate of the 'mean' synaptic cleft noradrenaline concentration. 110 6.2 AIMS OF THE STUDY 6.2.1. To determine whether infusions of the indirectly acting sympathomimetic tyramine produces a quantitative and reproducible release of endogenous neuronal noradrenaline. 6.2.2. To establish the nature of the relationship between blood pressure and plasma noradrenaline following release of endogenous neuronal noradrenaline. 6.2.3. To determine whether small changes in the rate of noradrenaline release (sufficient to cause a modest increase in blood pressure of about 15-20 mmHg) would give rise to detectable changes plasma noradrenaline. 6.3 METHODS 6.3.1.Protocol Six male volunteers were studied. Their mean age was 27 years (range 22 - 35), and their mean weight was 69 kg (range 58 - 75). The general conduct of the study, and the preparations for blood pressure measurement, venous sampling and handling of blood samples were as previously described in chapter 5. The subjects underwent infusions of 1-noradrenaline and of tyramine, one or two weeks apart and in balanced order. On each 111 occasion, after 40 minutes of supine rest, three sets of baseline observations (blood pressure, heart rate and blood sample) were made a 5 minute intervals and the means of these data taken as the control values. 6.3.2 Infusions The subjects then received six incremental ten minute infusions of tyramine or noradrenaline at the following doses, in order to raise blood pressure by a maximum of 15-20 mmHg. 1-Noradrenaline: 0.01, 0.02, 0.03, 0.05, 0.07, 0.10 ug/kg/min Tyramine: 2.5, 5.0, 7.5, 10.00, 12.5, 15.0 ug/kg/min 6.3.3 Measurements Blood pressure, heart rate and plasma noradrenaline and adrenaline were measured at 4, 8 and 10 minutes after each infusion. 6.4 RESULTS 6.4.1 Haemodynamic changes During 1-noradrenaline infusions, both systolic and diastolic blood pressure rose equally with increasing doses: from 107+11 to 125+9 mmHg systolic (+18), and from 1 2 + p to 87+7 mmHg diastolic (+15). This is shown in figure 6-2. Tyramine 112 infusions also raised systolic blood pressure by a mean of 16 mmHg, from 104+10 to 121+16. However, unlike the noradrenaline infusions, there was no significant change in diastolic blood pressure (figure 6-3). The increase in systolic pressure with tyramine was dose-related (table 6.1). Reflex slowing of the heart rate occurred with both infusions, but this was greater during noradrenaline infusions with a reduction of 8 bpm, compared with a reduction of 3 bpm with tyramine (see tables 6.1 and 6.2). 6.4.2 Plasma catecholamines The two amines had markedly different effects on plasma noradrenaline levels. At the highest infused dose of 1-noradrenaline, plasma noradrenaline levels increased from 0.30 to 1.63 ng/ml. In contrast the highest dose of tyramine induced a very similar increase in systolic blood pressure, but only raised plasma noradrenaline levels from 0.32 to 0.50ng/ml (figure 6-4). During both infusions, the changes in mean plama noradrenaline first achieved statistical significance when they exceeded baseline values by 40-50%. 6.4.3 Relationship between plasma noradrenaline and blood pressure The changes in blood pressure and plasma noradrenaline during tyramine infusion have been plotted in three different 113 ways. When all the individual absolute values of systolic blood pressure during the tyramine infusions were plotted against the corresponding plasma noradrenaline levels, no simple relationship was apparent and the regression line had a negative slope with a correlation coefficient of r=0.13 which was not statistically significant (figure 6-5). However, individual increases in systolic blood pressure correlated modestly though significantly with absolute values of plasma noradrenaline (r=0.39, p<0.05; see figure 6-6). When the increases in systolic blood pressure were related to the corresponding increases in plasma noradrenaline, the correlation was closer (r=0.54, p<0.001; see figure 6-7) The most striking correlations were obtained by analysis of the mean group data obtained at each dose during the two infusions. The rise in systolic blood pressure correlated closely with the plasma noradrenaline during both tyramine (r=0.98, p<0.001) and noradrenaline infusions (r=0.98, p<0.001) although the slopes of these relationships differed. The slope of the regression line for tyramine infusions was steep with a value of 80, whereas for noradrenaline infusions the slope of the regression line was flatter with a value of 13 (figure 6-8). A similar analysis with mean group changes in diastolic blood pressure during noradrenaline infusions gave a regression line with a slope of 8.4 (r=0.96; p< 0.001). Diastolic blood pressure did not rise significantly with tyramine. 114 6.4.4 Noradrenaline plasma kinetics Noradrenaline plasma clearance for individual subjects were calculated for plasma noradrenaline levels during 1-noradrenaline infusions. Clearance values varied widely between individuals, from 25 to 294 ml/kl/minute. The mean clearance for the group, however, remained approximately constant over the range of infused noradrenaline doses. This is shown in table 6.3. The calculated endogenous noradrenaline spillover rates are shown in table 6.4. This shows that noradrenaline spillover rates varied widely between individuals and that the plasma noradrenaline levels did not always reflect the apparent rate of noradrenaline release. 115 Table 6.1 Effects of tyramine on blood pressure, heart rate and plasma noradenaline levels (mean+SD) tyramine dose (uq/kq/min) control 2.5 5.0 7.5 10 12.5 15 Systolic BP 104 106 109 111* 113* ± 2 . 9 ** 121* (mmHg) 10 12 13 14 14 16 16 Diastolic BP 63 69 70 70 71 72 72 (mmHg) 7 5 6 7 6 6 7 Heart rate 58 59 58 57 57 55* 55* (bpm) 8 8 7 6 5 5 4 Plasma NA 0.32 0.31 0.37 0.39 0.44* 0.48* 0.50 (ng/ml) 0.10 0.08 0.11 0.13 0.11 0.13 0.19 *p<0.05, **p<0.01 compared with control Table 6.2 Effects of 1-noradrenaline on blood pressure, heart rate and plasma norarenaline (mean+SD) 1-noradrenaline (ug/kq/min) control 0.01 0.02 0.03 0.05 0.07 0.10 Systolic BP 107 108 109 112* 115** 117** 125** (mmHg) 11 11 10 12 11 11 9 Diastolic BP 72 76 76* 78* 81** 84** 87** (mmHg) 5 6 4 5 4 6 7 Heart rate 61 59 56* 56* 56* 54** 53** (bpm) 4 4 4 5 4 4 5 Plasma NA 0.30 0.43 0.56* 0.71* 1.03** 1.30* 1.63* (ng/ml) 0.11 0.19 0.24 0.36 0.55 0.82 1.19 *p<0.05, **p<0.01 compared with control 116 Table 6.3 Noradrenaline plasma clearance rates (ml/kg/min) during infusions of 1-noradrenaline. 1-noradrenaline (ug/kg/min) Subject 0.01 o O CN 0.03 0.05 0.07 0.10 AS 25 31 30 29 32 29 RC 77 62 125 98 125 125 MM 111 83 50 61 ? 56 DM 200 133 130 95 90 120 PI 200 222 120 119 205 294 MW 143 167 167 151 106 166 mean 126 116 104 92 117 132 sd 69 71 52 43 63 86 Table 6.4 Resting values for the spillover, clearance and plasma levels of noradrenaline at rest in six subjects. AS RC MM DM PI MW NA spillover 0.55 1.30 1.86 5.60 6.01 1.36 NA clearance 25 77 111 200 200 143 Plasma NA 0.37 0.24 0.24 0.40 0.41 0.13 units: NA spillover in ug/min. Plasma NA in ng/ml. NA clearance in ml/kg/min. 117 HO c h 2- c h 2- n h 2 tyramine noradrenaline Figure 6-1. Structure of tyramine and noradrenaline. 140 n systolic diastolic heart rate (bpm) 0.00 0.02 0.04 0.06 0.08 0.10 l-noradrenaline (ug/kg/min) Figure 6-2. The effects of l-noradrenaline on blood pressure and heart rate. 119 iue 3 Teefcs ftrmn nbodpesr and pressure rate. blood on heart tyramine of effects The 3. 6 Figure blood pressure ( m m Hg) 0 . 5075 1 12.5 5.0 10 7.5 000 2.55 tyramine (ug/kg/min) 120 iue -. h fet o yaieadlnrdeaie in l-noradrenaline, and tyramine of effects The 6-4. Figure approximately equipressor dose ranges, on plasma noradrenaline plasma on ranges, dose equipressor approximately concentrations. in E CL noradrenaline (ng/ml) 2.51 2 1.5- 1.0- .i i o.oi— . 0 5 10 5 0 ------(ug/kg/min) tyramine 1 ------1 r — 121 . 0.02 0.1 0.080.040.06 0.0 0 0 1 —1 « • 1 i 1— i •— i « i 1— i— 1— 1 l-noradrenaline (ug/kg/min) 150 -i r=0.13 03 X E 140 - E, 0) 130 - 13 03 03 CD iQ. _ T3 _oO _Q O 03>% 03 90 0.0 0.2 0.4 0.6 0.8 plasma noradrenaline (ng/ml) Figure 6-5. In this figure, all the plasma noradrenaline levels have been plotted against the corresponding blood pressure values; there is no correlation bewteen these sets of data. 122 40 n r=0.39 -10 H---- 1---- 1----1---- 1---- 1---- 1---- '---- 1 0.0 0.2 0.4 0.6 0.8 plasma noradrenaline (ng/ml) Figure 6-6. Plasma noradrenaline levels have been plotted against the corresponding increases in systolic blood pressure in order to reduce variability due to differences in resting blood pressures. There is a modest correlation, r=0.39. 123 Figure 6-7. Baseline variability in both blood pressure and plasma noradrenaline has been reduced further by plotting the increase in plasma noradrenaline against the corresponding increase in systolic blood pressure. This gives a closer correlation, r=0.54. 124 20 -1 o> X E 15 - E C l tyramine CD o 2 00 10 - CO>% plasma noradrenaline (ng/ml) Figure 6-8. This shows mean data for the whole group. The mean plasma noradrenaline has been plotted against the mean increases in systolic blood pressure. This shows close and highly significant correlations during both tyramine (p<0.01) and 1-noradrenaline (p<0.01) infusions. 125 6.5 DISCUSSION 6.5.1 Tyramine Although it is well established that tyramine releases noradrenaline from sympathetic nerve endings, its mechanism of action is not fully understood (Rapoport et al, 1981). It is currently envisaged that tyramine is actively transported into sympathetic nerve endings by the neuronal reuptake mechanism; it is also a competitive inhibitor of noradrenaline uptake (Burgen and Iversen, 1965). Subsequently, active incorporation of tyramine into adrenergic storage vesicles causes displacement of noradrenaline into the cytosol. This free cytosolic noradrenaline is a substrate of mitochondrial monoamine oxidase (Leitz and Stefano, 1971). It is clear however that the concentration of intact noradrenaline in the cytosol is sufficiently high in the presence of tyramine to accelerate noradrenaline efflux from the nerve terminal. These effects are independent of extracellular calcium ion concentration, are not inhibited by colchicine or neurotoxins and are not accompanied by the release of dopamine beta hydroxylase (Smith, 1973; Trendelenburg, 1979). Thus, the action of tyramine clearly differs from exocytotic release of noradrenaline secondary to nerve terminal depolarisation. However, two factors suggest that the use of tyramine might be valid for the purpose of these studies. 126 First, it releases noradrenaline into or close to the synaptic cleft, and it is reasonable to assume that the effects and subsequent fate of this noradrenaline would be comparable to that following physiological release. Second, the infusion rate can be adjusted to make relatively small changes in the rate of noradrenaline efflux from nerve endings. 6.5.2 Relationship between blood pressure and plasma noradrenaline The results of this study show that tyramine infusions induced dose-related increases in systolic blood pressure, as well as small but significant dose-related increases in plasma noradrenaline levels. The smallest increase in the mean plasma noradrenaline that achieved statistical significance was 0.44 ng/ml - an increase of 37% over the baseline value of 0.32 ng/ml. This was accompanied by a 10 mmHg rise in blood pressure. Thus, even small rises in the rate of noradrenaline release (sufficient in this instance to cause a 10 mmHg rise in blood pressure) can be associated with small but just-detectable increases in plasma noradrenaline, at least under the conditions of this study. However, the relevant conditions were virtually ideal: namely, mean, within-group data obtained during an acute study, in a controlled laboratory environment, with medical and scientific personnel as subjects. In order to determine whether these findings are more widely applicable, eg to comparative studies between groups, the 127 data obtained during the tyramine infusions were plotted in three different ways to investigate the relationship between plasma noradrenaline and the blood pressure response. In the plot of all individual data points (in figure 6-6), there was no correlation between the absolute values of systolic pressure and plasma noradrenaline. The absence of a plamsa NA /SBP relationship is explained by two factors. The resting values for systolic BP varied from 96 to 119 mmHg, and for plasma noradrenaline from 0.16 to 0.47 ng/ml; however, due to the wide range of clearance rates there was no relationship between these two parameters. Furthermore, the response of both blood pressure and plasma noradrenaline to tyramine in individual subjects was variable. A different picture emerged when individual changes in systolic blood pressure (ie 'delta' SBP) were considered. The plot of increase in SBP against plasma noradrenaline shown in figure 6-6 corrected for interindividual differences in baseline blood pressure, and this showed a significant correlation. A still closer correlation was obtained when increases in SBP were plotted against increases in plasma noradrenaline to correct for the differences in resting noradrenaline as well. Finally, by analysis of the mean increases in SBP for the entire group, variability both among and within individuals was further reduced. This plot showed a highly significant correlation between the mean plama noradrenaline and the mean 128 'delta' SBP obtained at each dose of tyramine. Ths slope of this relationship was steep and showed that a rise in systolic blood pressure of lOmmHg was associated with a rise in plasma noradrenaline of 0.12 ng/ml. These data suggest that (a) a quantitative relationship exists between sympathetic nervous activity and plamsa noradrenaline concentrations, and (b) that small changes in sympathetic activity can be detected in highly controlled conditions. However, such is the variability between individuals that differences in sympathetic activity will only account for a small proportion of the total variance observed in plasma noradrenaline levels. It follows that the usefulness of plasma noradrenaline to compare sympathetic activity between individuals is limited, particularly when the differences are small. On the other hand, for acute studies performed within a homogeneous group, plasma noradrenaline can give a useful indication of sympathetic activity. 6.5.3 Possible differences in in the effects of endogenous noradrenaline and infused 1-noradrenaline. The results of the 1-noradrenaline infusions were in accordance with the well known actions of this catecholamine. The high plama noradrenaline levels achieved during these infusions contrasted with the small increase in plasma noradrenaline seen during tyramine infusions; this is explained by the dilution in the blood stream of the relatively small 129 spillover of noradrenaline released into the synapse by tyramine. It was surprising to find such a marked difference between the slopes of the NA/SBP curves for the two amines. In part, this difference may reflect the gradient between synaptic cleft noradrenaline concentrations and plasma noradrenaline concentrations required to produce a given increment in blood pressure. However, had the exogenous and endogenous noradrenaline stimulated the identical receptor population during both infusions, a parallel shift between two curves of similar slope would be expected. That this was not the case raises the possibility that the pressor mechanisms of exogenous infused 1-noradrenaline and endogenous tryamine-released noradrenaline differ. In addition, 1-noradrenaline raised diastolic pressure, but tyramine did not. Several explanations are possible. Tyramine may release noradrenaline from sympathetic neurones and peripheral blood vessel walls causing stimulation of both alpha and beta receptors. The mutually antagonistic effects of the two receptor types could account for the negligible increase in diastolic blood pressure. It would be likely in this case that the preferential increase in systolic blood pressure with tyramine resulted predominantly from the inotropic effect of intracardic noradrenaline release. On the other hand, there is evidence for the 130 existence of vasoconstrictor alpha-2 receptors located extrasynaptically in vascular smooth muscle. It is likely that exogenous noradrenaline would stimulate these receptors as well as postsynaptic alpha-1 receptors (Langer, Massingham et al, 1980; Yamaguchi and Kopin, 1980). This would constitute a different receptor population from that stimulated (indirectly) by tyramine, and could account for the non-parallel curves. 6.5.4 Comparison with other studies. Because the commonly-used sympathetic stimuli involve marked changes in sympathetic activity and/or central haemodynamics, it is difficult to obtain a quantitative comparison of these results with other published data. However, it is interesting to note the finding of Cryer et al (1976), that during endogenous noradrenaline release by cigarette smoking there was a mean rise in supine blood pressure of 10 mmHg, and this was associated with a O.lng/ml increase in plasma noradrenaline. This is in close agreement with the results obtained with tyramine - 0.12 ng/ml for a 10 mmHg rise in blood pressure. 6.6 SUMMARY This study showed that infusions of tyramine can be used successfully to cause the release of endogenous neuronal noradrenaline, with effects on blood pressure and plasma 131 noradrenaline levels analogous to those of sympathetic nervous system activity. This model was used to study the effects of small incremental changes in the rate of neuronal noradrenaline release, and these were compared with the effects of exogenous 1-noradrenaline. Tyramine caused dose-related increases in systolic blood pressure and plasma noradrenaline. A highly significant quatitative relationship between blood pressure and plasma noradrenaline was evident for the group as a whole, and it was possible to detect relatively small changes in noradrenaline release. Considerable variability was observed in noradrenaline clearance; this limits the ability of plasma noradrenaline to compare sympathetic activity between indivduals, but it may still be a useful index of short-term changes of sympathetic activity within a group. 132 CHAPTER 7 THE EFFECTS OF PROPRANOLOL ON THE PRESSOR RESPONSE TO TYRAMINE 133 7.1 INTRODUCTION The validity of peripheral venous measurements of plasma noradrenaline as a biochemical index of sympathetic activity depends partly on the changes in sympathetic outflow being generalised. Changes in plasma noradrenaline resulting from non-uniform regional sympathetic discharges would be difficult to interpret. It is therefore necessary to determine whether the mechanism by which tyramine raised blood pressure was compatible with a generalised increase in cardiovascular sympathetic activity. The pressor mechanism of tyramine is not well described. It has often been assumed that the rise in blood pressure is related to an increase in peripheral resistance, but the data presented in chapter 6 suggested indirectly that an increase in cardiac contractility may also be important. This is not surprising, since tyramine is as likely to displace neuronal noradrenaline from sympathetic nerve terminals in the heart as elsewhere. A cardiac inotropic effect mediated by beta-1 adrenoceptors might also account for the differences in the NA/SBP relationship, for the pressor effect of infused exogenous 1-noradrenaline is due almost totally to peripheral alpha-mediated vasoconstriction. 7.2 AIMS OF THE STUDY 134 This study was undertaken to determine whether the mechanism of action of tyramine was consistent with a generalised increase in peripheral sympathetic activity (mediated by vasoconstrictor alpha receptors), as opposed to a predominantly cardiac beta-mediated effect. Beta-blockade should therefore reduce the effects (if any) of tyramine on cardiac contractility; furthermore, by blocking vasodilator beta-2 receptors, it should enhance tyramine-induced peripheral vasoconstriction through unopposed alpha-receptor effects. Thus, a nonspecific betablocker such as propranolol should antgonise the pressor action of tyramine if these are mediated principally through beta receptors; alternatively it may potentiate the increase in blood pressure if this is mediated by alpha-receptors. 7.3 METHODS 7.3.1 Subjects Five normotensive male volunteers (mean age 28+3 years; mean weight 69+6kg) were recruited from medical and laboratory staff. General conduct and conditions of the study were as previously described. 7.3.2 Protocol The study design was single blind. Incremental pressor 135 infusions of tyramine were administered on two occasions, following beta blockade or matching placebo. The subjects attended on two occasions one week apart, and received either oral propranolol 160mg or matching placebo. Intravenous cannulae, blood pressure recording and ECG electrodes were set up as previously described. After 60 minutes of quiet supine rest, two sets of baseline measurements were made 10 minutes apart; these were blood pressure, heart rate, blood samples for plasma catecholamines, and systolic time intervals. The mean of two results was taken as the control value. (Note: the results of the cardiac systolic time intervals will be presented separately in chapter 8). The subjects then received five incremental 20 minute infusions of tyramine in one of two dose ranges. The lower dose range was administered if they had received placebo: tyramine 2.5, 5.0, 7.5, 10.0 and 12.5ug/kg/min. The higher dose range was administered if they had received active propranolol 160mg: tyramine 5.0, 10.0, 15.0, 20.0 and 25.Oum/kg/min. The propranolol or placebo was given 2 hours before beginning the tyramine infusions. 7.3.4 Measurements Blood samples and readings of blood pressure, heart rate and systolic time intervals were taken after 15 minutes of each infusion. 136 7.4 RESULTS 7.4.1.Propranolol and baseline values The effects of tyramine infusions on blood pressure, heart rate and plasma noradrenaline after placebo are summarised^table 7.1 and after oral propranolol in table 7.2. The only signficant change induced by propranolol in the preinfusion baseline data was a small increase in the resting diastolic pressure: placebo 63+4 mmHg vs propranolol 68jf6 mmHg (p<0.05) . The resting systolic blood pressure and heart rate were slightly but not significantly lower on propranolol, indicating that the subjects' basal sympathetic tone after 60 minutes of rest was low. 7.4.2 Plasma noradrenaline levels Resting plasma noradrenaline levels were not altered by propranolol: 0.283+0.06 ng/ml after placebo, compared with 0.279+0.06 ng/ml after propranolol. Similarly, at the two tyramine doses that were common to both dose ranges, ie 5.0 and 10.Oug/kg/min, the plasma noradrenaline levels after propranolol and placebo were nearly identical, as shown in figure 7.1. This suggests that propranolol did not have any important effect on the neuronal uptake of tyramine, the release of noradrenaline, or on its subsequent disposition in the blood stream. 137 7.4.3 Systolic blood pressure During the control infusions (ie after placebo), systolic blood pressure rose from 114+5 to 138+6 mmHg at the highest tyramine dose of 12.5 ug/kg/min. Propranolol attenuated this response, so that the maximum systolic pressure was 133+18 mmHg despite the higher tyramine dose of 25 ug/kg/min. Figure 7-2. is a log dose-response plot showing the rise in blood pressure with increasing doses of tyramine after placebo and propranolol. Both curves have similar configurations, but the curve after propranolol is shifted to the right in an approximately parallel fashion. The data from each curve was replotted in figure 7-3 to show the relationship between increase in systolic blood pressure and the corresponding plasma noradrenaline levels. It is evident from this plot that propranolol caused a parallel shift of the delta NA/SBP curve to the right. 7.4.4 Diastolic blood pressure After placebo, diastolic blood pressure remained unchanged with increasing doses of tyramine: 63+4 mmHg preinfusion, and 62+3 mmHg at the highest tyramine dose. After propranolol, the diastolic blood pressure did not change at lower doses of tyramine (5.0 and 10.0 um/kg/min), but rose significantly at doses of 15.0 ug/kg/min and above, to a maximum of 82+9 mmHg at 138 the highest dose (Table 7.2). 7.4.5 The distinction between alpha and beta mediated effects Some of the data from tables 7.1 and 7.2 have been replotted to assess the differential effects of tyramine on alpha and beta receptors. In order to do this, it has been assumed that the tyramine-induced rise in systolic blood pressure (after placebo) reflects chiefly a beta receptor-mediated increase in cardiac contractility. The rise in disatolic blood pressure after propranolol was taken as reflecting alpha receptor mediated vasoconstriction. This is shown in figure 7-4, which is a log-dose response plot showing "alpha" and "beta"-mediated increases in blood pressure with increase in tyramine dose. This shows that tyramine indirectly stimulates both alpha and beta-receptors, and that the beta-mediated response is detectable at lower doses. When these alpha and beta effects are related to plasma noradrenaline levels, it is clear that the beta effects are associated with smaller increments in plasma noradrenaline (figure 7-5). 139 Table 7.1 Effects of tyramine after placebo (mean+SD) Tyramine (ug/kg/min) control 2.5 5.0 7.5 10.0 12.5 Systolic BP 114 115 116 121* 132** 138** (mmHg) 5 10 8 9 4 6 Diastolic BP 63 62 64 63 62 62 (mmHg) 4 5 4 5 7 3 Heart rate 62 62 61 60 60 60 (bpm) 9 7 6 7 10 10 Plasma NA 0.28 0.28 0.35 0.42** 0.44* 0.46** (ng/ml) 0.06 0.07 0.06 0.04 0.10 0.10 *p<0.05, **p<0.01 compared with baseline control 140 Table 7.2 Effects of tyramine after propranolol 160 mg po (mean+SD) Tyramine (ug/kg/min) control 5.0 10.0 15.0 20.0 25.0 Systolic BP 111 112 114 120* 130* 134* (mmHg) 8 8 9 11 20 18 Diastolic BP 68 69 70 74* 77 82* (mmHg) 6 5 8 4 9 9 Heart rate 60 58 56 54* 54** 56 (bpm) 7 8 6 5 6 9 Plasma NA 0.28 0.36 0.45* 0.54** 0.59* 0.63** (ng/ml) 0.06 0.07 0.09 0.10 0.20 0.12 *p<0.05, **p<0.01 compared with baseline control 141 iue71 Theofeffectstyramineonnoradrenaline plasma 7-1.Figure concentrations inandconcentrationsthepresenceofabsence propranolol. plasma noradrenaline (ng/ml) 142 -10 *t " i i i i i i i r i i 1 3 10 30 tyramine (ug/kg/min) Figure 7-2. Log dose-response plot showing the effect of propranolol on the pressor response to tyramine. The tyramine dose-response curve has been shifted to the right in an approximately parallel manner. 143 Figure 7-3. The relationship between plasma noradrenaline and increased systolic blood pressure during tyramine infusions, before and after propranolol. Propranolol has shifted the curve to the right. 144 O) X E E 3 CO Figure 7-4. The dose-response relationship between tyramine, and alpha and beta-mediated pressor responses. The 'alpha' response was the increase in diastolic pressure during tyramine infusions after propranolol; the 'beta' response was the increase in systolic pressure after tyramine alone. 145 30 n ___ O) 25 ■ X E E 20 CD (/)"5 in 15 CD CL ■O O 10 O JO c CD 5 C/3ro CD 0c 0 -5 plasma noradrenaline (ng/ml) Figure 7-5. The re la tio n s h ip between alpha and beta-m ediated pressor effects of tyramine and plasma noradrenaline levels. The 'alpha' effects are associated with higher plasma noradrenaline levels. (As in the previous figure,the 'alpha' response was the increase in diastolic pressure during tyramine infusions after propranolol; the 'beta' response was the increase insystolic pressure after tyramine alone.) 7.5 DISCUSSION 7.5.1 Effects of propranolol on systolic pressure. The effects of the tyramine infusions (after placebo) confirmed the observations made in the previous study. This showed a dose dependent increase in systolic blood pressure which was associated with a small but significant rise in plasma noradrenaline. The diastolic blood pressure remained unchanged. However, propranolol altered both the magnitude and pattern of this response. The rise in systolic pressure was markedly reduced, with a parallel rightward shift in the log tyramine-systolic blood pressure curve. This suggests that an important component of the pressor response to tyramine is mediated by beta receptors, and these are presumably cardiac. This would be in keeping with the haemodynamic pattern of a predominant increase in the pulse pressure with little change in the diastolic blood pressure. Thus it is likely that the pressor effect of tyramine is related chiefly to increased cardiac contractility and output rather than to increased peripheral vascular resistance. 7.5.2 Effects of propranolol on diastolic pressure. The diastolic blood pressure response was also influenced by beta blockade. After propranolol the diastolic pressure rose significantly with the higher doses of tyramine (15 ug/kg/min and over), whereas with lower doses there was no rise, 147 irrespective of whether propranolol had been given or not. It is therefore uncertain whether the rise in diastolic blood pressure was a consequence solely of unmasking alpha-mediated vasoconstriction, or whether it would have risen anyway with higher doses of tyramine. However, it seems unlikely that there were significant vasodilator beta effects in this dose range. 7.5.3 Relative importance of alpha and beta responses. The rise in diastolic pressure after beta blockade provided an opportunity to assess alpha-mediated responses to tyramine, as it were, in isolation. Of course it is recognised that such a distinction is largely artificial but does allow a means of comparing alpha and beta receptor mediated responses, and their relationship to tyramine dose and to plasma noradrenaline. Thus, it appears that tyramine indirectly stimulates both types of receptor. The beta-mediated effects are seen at lower doses of tyramine and are associated with a small increment in plasma noradrenaline. By contrast, the alpha-mediated pressor effects are seen at high doses of tyramine, and are associated with a much greater increment in plasma noradrenaline. These relatively high noradrenaline levels were similar to those seen in the previous study with infused exogenous noradrenaline, and suggest that high plasma (as opposed to intrasynaptic) noradrenaline concentrations may be important for the vasoconstriction seen with higher tyramine doses. Furthermore, differences in cardiac and arteriolar synaptic cleft widths, nerve terminal densities and post 148 synaptic receptor densities may also be important (Bevan, 1977). 7.6 SUMMARY This study was undertaken to determine whether tyramine raised blood pressure by causing peripheral arteriolar vasoconstriction, or by stimulation of cardiac contractility. This was performed by assessing the effects of propranolol on the pressor response to tyramine. Propranolol attenuated the effects of tyramine, suggesting that its principal mechanism of action was though stimulation of cardiac contractility, probably by intracardiac noradrenaline release. Tyramine also (indirectly) stimulates peripheral alpha receptors, although higher doses are required. 149 CHAPTER 8 THE EFFECTS OF TYRAMINE ON CARDIAC SYSTOLIC TIME INTERVALS BEFORE AND AFTER PROPRANOLOL 150 8.1 INTRODUCTION An assessment of the effects of tyramine on the systolic time intervals before and after propranolol was performed as part of the study reported in chapter 7. The results will be described in this chapter since the data merit separate discussion. These data were obtained as an additional and more direct means of determining the mechanism of action of tyramine. Measurement of the systolic time intervals is a method of assessing left ventricular function in terms of well defined events in the cardiac cycle. Since it is noninvasive it can be performed repeatedly, and is therefore well suited to the study of drugs with cardiovascular effects. The interpretation and significance of the systolic time intervals (3TI) has been extensively validated (Lewis et al, 1977) . The particular attraction of this technique in assessing the action of tyramine lie in its simplicity: drugs which increase peripheral resistance tend to prolong the STIs; whereas drugs with positive inotropic effects shorten them. 8.2 AIMS OF THE STUDY The purpose of this study was to test the hypothesis that the pressor action of tyramine is due to a cardiac positive 151 inotropic effect. In addition, the relationship between plasma noradrenaline and cardiac sympathetic activity was assessed. 8.2 METHODS The subjects who took part in this study, and the experimental protocol have been described in chapter 7. The systolic time intervals electromechanical systole (QS2), left ventricular ejection time (LVET) , and pre-ejection period (PEP) were measured using the methods of Weissler et al (1968) which have been described in chapter 5, section 5.6.3. All STI data are reported as the index - ie corrected for heart rate, as summarised in table 5.1 (section 5.6.3). 8.3 RESULTS 8.3.1 Effects of tyramine The effects of tyramine on the rate-corrected systolic time intervals are shown in table 8-1. In all subjects all intervals were progressively and consistently shortened by increasing doses of tyramine. QS2I fell by 12%, from 512+11 ms to 462+15 ms. LVETI fell by 8%, from 394+8 ms to 362+12 ms. PEPI fell by 18%, from 118+6 ms to 99jfl0 ms. 152 8.3.2 Effects of propranolol on resting STI Propranolol tended to lengthen the resting systolic time intervals, but this was only statistically significant for QS2I: 512tll ms after placebo, compared with 526+17 after propranolol (p<0.05). PEP was lengthened from 118+6 ms to 127+7 ms (ns). LVETI was almost unchanged at 394+9 ms after placebo, and 399+16 ms after propranolol. 8.3.3 Tyramine infusions after propranolol Propranolol almost completely abolished the shortening of all systolic time intervals during the infusion of tyramine. These results are shown in table 8.2, and in figures 8-1, 8-2 and 8-3. 8.3.4 Relationship between systolic time intervals and blood pressure During the tyramine infusions (after placebo), all systolic time intervals showed a consistent inverse relationship with systolic blood pressure. This was true of the mean group data, as shown in figures 8-4, 8-5 and 8-6/ it was also clearly the case when each individual's data was inspected. This is shown in figure 8-7. After propranolol, the changes in systolic time intervals were small and inconsistent. There was now no obvious 153 relationship between these data and the changes in blood pressure. 8.3.4 Systolic time intervals and plasma noradrenaline When all the individual data points for all three systolic time intervals were plotted against the corresponding plasma noradrenaline values, modest but statistically significant correlations were obtained for QS2I (r=0.55, P<0.001) and for LVETI (r=0.63, P<0.001), but not for PEPI (r=0.35, ns). However, for the group as a whole, highly significant linear correlations were obtained between the changes in systolic time intervals and the changes in plasma noradrenaline. This is shown in figures 8-8. After propranolol, no significant correlations were obtained between changes in the systolic time intervals and changes in plasma noradrenaline. 154 Table 8.1 Effects of tyramine on the cardiac systolic time intervals, systolic pressure, and plasma noradrenaline (mean+SD) Tyramine (uq/kg/min) control 2.5 5.0 7.5 1 0 . 0 1 2 .5 QS2 index 512 512 502 489** 472** 462** (msec) 11 20 24 20 10 15 LVET index 394 396 385 376** 371** 363** (msec) 8 9 15 12 4 12 PEP index 118 117 116 113 1 0 1 * 9 9 ** (msec) 6 5 12 6 9 10 PEP/LVET 0.03 0.03 0.03 0.03 0.27* 0.27 0 . 0 1 0 . 0 1 0.03 0 . 0 2 0 . 0 2 0.03 Systolic BP 114 .115 116 1 2 1 * 132** 138** (mmHg) 5 10 8 9 4 6 Plasma NA 0.28 0.28 0.35 0.42** 0.44* 0.46** (ng/ml) 0.06 0.07 0.06 0.04 0 . 1 0 0 . 1 0 *p<0.05, **p<0.01 compared with (baseline) control 155 Table 8.2 Effects of tyramine after propranolol 160 mg po (mean+SD) Tyramine (ug/kg/min) control 5.0 1 0 . 0 15.0 2 0 . 0 25.0 QS2 index 526+ 520* 515* 512* 515 518 (msec) 17 17 19 18 14 18 LVET index 399 396 391 386 391 391 (msec) 16 17 21 20 15 22 PEP index 127 124 127 126 124 127 (msec) 7 8 3 9 8 9 PEP/LVET 0.32 0.31 0.33 0.33 0.32 0.33 0 . 0 2 0.03 0 . 0 2 0.04 0 . 0 2 0.04 Systolic BP 1 1 1 1 1 2 114 1 2 0 * 130* 134* (mmHg) 8 8 9 11 20 18 Plasma NA 0.28 0.36 0.45* 0.54** 0.59* 0.63** (ng/ml) 0.06 0.07 0.09 0 . 1 0 0 . 2 0 0 . 1 2 +p<0.05 compared with control study (before propranolol) *p<0.05, **p<0.01 compared with (baseline) control 156 Figure 8-1. The effects of tyramine on the preejection period (PEP) index, before and after propranolol. 157 tyramine (ug/kg/min) Figure 8-2. The effects of tyramine on the left ventricular ejection time (LVET) index, before and after propranolol. 158 tyramine (ug/kg/min) Figure 8-3. The effects of tyramine on the Q to S2 interval (QS2) index, before and after propranolol. 159 Figure 8-4. Relationship between the preejection period (PEP) index and systolic blood pressure. 160 v Figure 8-5. Relationship between the left ventricular ejection time (LVET) index, and systolic blood pressure. 161 Figure 8 -6 . Relationship between the Q to S2 interval (QS2) index, and systolic blood pressure. 162 QS2 index (ms) Figure 8-7. The relationship between the rise in systolic blood pressure and the shortening of the Q to S2 interval, shown in all 5 individual subjects. 163 Figure 8-8. Correlation between plasma noradrenaline levels and the systolic time intervals. 164 8.4 DISCUSSION 8.4.1 Determinants of the systolic time intervals Measurement of systolic time intervals is a well established non-invasive technique for the study of cardiac systolic function. Although its role in clinical investigation has been superceded by newer methods, it remains a useful tool for assessing the effects of drugs on the heart (Gibson, 1978). The directly measured intervals are QS2 and LVET. Due to the transmission delay in registration of the carotid pulse wave, usually 15 to 20 ms, the PEP is not measured directly but obtained by subtracting LVET from the QS2. The PEP itself can be further subdivided into two components. These are the preisovolumic period which is relatively constant and consists principally of time required for intracardiac electrical conduction. The other sub-interval is the isovolumic contraction time which can change markedly in response to physiological and pharmacological stimuli. Changes in PEP are therefore largely due to changes in isovolumic contractions (Lewis et al, 1977; Nandi and Spodick, 1977). Systolic time intervals are affected by a number of factors, (a) Heart rate - This was recognised as a major determinant of ejection time by Garrod in 1874. Empirical OSl regression equationsAcomonly used to correct the measurement for the effects of heart rate, and this consists of expressing values as those predicted to occur if a heart rate were 165 extrapolated to zero. The commonly used regression equations were those published by Weissler and are shown in table 5-1. Both QS2 and LVET shorten appreciably with faster heart rates, but the effect on PEP is marginal. (b) Inotropic state - Changes in inotropic state (independent of left ventricular filling) alter the systolic time intervals in a predictable direction. Positive inotropic agents consistently shorten all intervals. The degree of shortening has been shown to be related to a variety of invasive indices of contractility including peak LV dp/dt and Vmax (Ahmed et al, 1972). Of negatively inotropic drugs, the group most studied has been the beta blockers. The exact effects of these drugs on systolic time intervals depends on a number of factors including subject's prevailing level of sympathetic tone. In general the PEP is lengthened while the QS2 and LVET may be unchanged or shortened depending on dose. (c) Peripheral resistance - Increased peripheral resistance increases the time required for the left ventricular stroke volume to be ejected, and it therefore prolongs all systolic time intervals. A reduction in peripheral resistance, eg by vasodilator drugs shortens all intervals by the opposite effect. (d) Left ventricular filling - The effects of changes in left ventricular filling are unpredictable because they are complex and often accompanied by other changes in central 166 haemodynamics which may result in a net change opposite to what had been anticipated. For example, an increase in the left ventricular diastolic volume (prior to an increase in stroke volume) might prolong ejection time whereas by Starling's Law the myocardial contractility would increase and this could shorten the ejection time (Gibson, 1978). 8.4.2 Effects of tyramine and propranolol In this study, the consistent dose-related shortening of all intervals by tyramine and the abolition of this response by propranolol, exactly parallel the effects on blood pressure reported in chapter 7. These results provide convincing evidence that tyramine has important positive inotropic effects. Furthermore, since the unchanged diastolic pressure suggests that there were no major changes in peripheral resistance, it is likely that these inotrophic effects were largely responsible for the rise in systolic blood pressure. This view is strongly supported by the close relationship between the progressive shortening of all intervals and the progressive increase in systolic blood pressure. Had the pressor effect resulted from an increase in peripheral resistance, the systolic time intervals would have been lengthened rather than shortened. Comparison of these data to those reported by Walsh et al (1982) shows that the maximal effects of tyramine (1 2 .5ug/kg/min after placebo) were similar to between 5 and lOug/kg/min of dobutamine, a beta-1 agonist. 167 The resting intervals in this study were all noted to be approximately 6-7% shorter than normal data reported in other studies (Lewis et al, 1977). This may be partly because the subjects of this study were uniformly young compared with the much wider age range of the patients studied by Weissler's group. Another possible explanation would be that the paper speed of the recording device was not exactly lOOmm/sec. However, neither of these factors would invalidate the conclusions of this study, since the decisive change in STI was qualitative rather than quantitative. This study also demonstrated close correlations between changes in systolic time intervals and plasma noradrenaline. How these should be interpreted in the light of the predominantly cardiac pressor effects of tyramine is not clear, since tyramine appears to have predominantly regional rather than generalised effects on neuronal noradrenaline release. It may be that the peripheral effects of tyramine in releasing neuronal noradrenaline are a good reflection of its intracardiac effects; on the other hand it may have been a purely incidental association. This cannot be determined from the current data, and further studies are required to assess the value and significance of plasma noradrenaline in the presence of regional rather than generalised changes in sympathetic activity. 168 8.5 CONCLUSIONS Cardiac systolic time intervals were measured as a noninvasive means of investigating the mechanism of action of tyramine. The results showed that tyramine had cardiac inotropic effects that that closely parallelled its pressor effects. Both were abolished by propranolol. It is concluded that tyramine raises blood pressure by stimulation of cardiac contractility. The significance of increases in plasma noradrenaline when sympathetic activity is regional rather than generalised is uncertain. 169 CHAPTER 9 TYRAMINE AND COLD STRESS 170 9.1 INTRODUCTION The evidence that tyramine acts principally by stimulation of cardiac contractility forces a reappraisal of the data concerning the tyramine-induced changes in blood pressure and in plasma noradrenaline. Even though the rises in plasma noradrenaline were small, it is very unlikely that they resulted solely from an increase in cardiac noradrenaline release since the heart contributes only about 2 % of the circulating pool (Brown et al, 1981). A proportion of the circulating noradrenaline may have been released from peripheral resistance vessels, for alpha receptor-mediated vasoconstrictor effects were seen at hgigher doses of tyramine. Whether there was also some contribution from veins and from non-cardiovascular tissues is unknown. However, if the rise in plasma noradrenaline reflects release from a number of different sources, then the significance of any apparent relationship between this and blood pressure must be in doubt. The effects of tyramine might be better understood if compared to a physiological stimulus of sympathetic activity, where changes in plasma noradrenaline might be expected to reflect release only from tissues relevant to the cardiovascular effects. Many conventional stimuli of sympathetic activity would not be suitable for this, being too intense, too brief (insufficient time for equilibration of noradrenaline in plasma), or for causing adrenaline release as well. 171 However, exposure to cold ambient temperatures is known to cause a modest sympathetic activation associated with an increase in circulating noradrenaline. There is also a rise in blood pressure which is associated with increased systemic resistance due to arteriolar constriction. The purpose of this study was to compare the effects of tyramine and cold stress on changes in blood pressure, heart rate and PNA. In particular whether these two clearly different pressor mechanisms would give similar changes in PNA for the same rise in blood pressure. 9.2 METHODS 9.2.1 Subjects Six healthy male volunteers (mean age 27+4 years; mean weight 66+4 kg) had tyramine infusions or cold stress studies 2 weeks apart and in balanced order. Tobacco, alcohol and caffeine were avoided for 12 hours beforehand. 9.2.2 Tyramine infusions These took place in a clinical laboratory and began at 0900 hours with the subjects supine throughout. The general conduct of the studies was as prevously described. After 40 minutes supine rest, 3 sets of basal observations (BP, heart 172 rate and blood sampling for noradrenaline and adrenaline) were made at minute 5 intervals, the mean results being taken as the control. The subjects then received 3 incremental 30 minute infusions of tyramine at 5.0, 10.0 and 15.0 ug/kg/min; blood pressure and heart rate were recorded, and blood samples taken at 10,20 and 30 minutes during each dose. 9.2.3 Cold stress studies Began at 0900 and took place in a laboratory cold room maintained at 4-5oC. The subjects lay on a mattress placed on the floor and were warmly covered with blankets. Blood samples were drawn via an indwelling 19 gauge butterfly needle inserted under local anaesthesia into an antecubital vein, and kept patent with heparinised 0.9% NaCl. After 40 minutes warm supine rest, BP, HR and blood samples for catecholamines were taken twice, 5 minutes apart, and the mean of these results was taken as the control. The subjects, who wore only shorts or lightweight trousers, were then abruptly exposed to the cold air temperature by removal of their blankets. Cold stress continued for 30 minutes; BP, HR and blood samples were taken at 5, 10, 15, 20, 25 and 30 minutes. The subjects were closely observed throughout: there was no overt shivering although superficial fasciculation was observed in some subjects. 9.2.4 Measurements 173 In all studies, blood pressure was measured in triplicate by a semi-automatic sphygmomanometer (Roche Arteriosonde). During all tyramine infusions the electrocardiogram was continuously monitored, and the heart rate was measured from the electrocardiographic recording. In the cold stress studies the heart rate was obtained by counting the radial pulse. Blood samples for estimation of noradrenaline and adrenaline were handled, stored and assayed as previously described. 9.3 RESULTS 9.3.1 Effects of tyramine The effects of tyramine infusions on blood pressure, heart rate and plasma catecholamines are shown in table 9.1. Systolic pressure rose gradually with increasing tyramine dose, and this was accompanied by a small but significant fall in the heart rate. Diastolic pressure was unchanged (figure 9-1). Plasma noradrenaline levels also rose gradually with increasing tyramine dose. It should be noted that the mean control plasma noradrenaline level of 0.547ng/ml was higher than the value of 0.3-0.4 ng/ml usually recorded in this laboratory for normal subjects at rest. This is because two of the subjects had high-normal resting noradrenaline values (of 0.690 and 0.779 ng/ml respectively) during both the tyramine infusion and the cold stress study, although their responses were similar to those of the other subjects. Plasma adrenaline levels did not 174 increase during the infusions (figure 9-2). 9.3.2 Effects of cold stress The effects of cold stress on blood pressure, heart rate and plasma catecholamines are shown in table 9.2. Systolic pressure rose significantly (although by less than during the tyramine infusions); the most rapid increase occurred during the first 15 minutes of cold stress, with no further increase in the last 15 minutes (figure 9-3). Similarly, plasma noradrenaline levels rose during the first 20 minutes of cold stress and were essentially unchanged during the final 10 minutes. Neither heart rate nor plasma adrenaline levels changed, but in contrast with the tyramine infusions, diastolic BP rose progressively and significantly during cold stress. 9.3.3 Relationship Between Blood Pressure and Plasma Noradrenaline In figure 9-5 the blood pressure and noradrenaline data from tables 9.1 and 9.2 have been plotted to show the relationships between mean plasma NE concentrations and mean increases in systolic pressure obtained during tyramine infusions and cold stress. This shows that plasma noradrenaline is closely correlated with increases in systolic pressure during release of endogenous noradrenaline both by tyramine (p < 0.01) and by cold stress (p < 0.01). It is also clear, however, that these relationships differ quantitately: the slope of the 175 regression line for tyramine infusions is 105, and this is markedly steeper than the slope of 17 obtained for the cold stress studies. Tyramine infusions were not associated with increased diastolic pressure, and there was no relationship between the plasma noradrenaline and diastolic pressure changes. By contrast, the increases in diastolic pressure during cold stress were closely correlated with the noradrenaline with a slope similar to that obtained for systolic pressure. 176 * * * * Mf O oo m £ ■** ON 177 G * * •x c/a •iH * * OV vO <}• •rt * * oo cn VO Mj- 4J E vO Ov o m in ov ov cn id T“H oo o o O in rH m CM o o o o id 6 CO id f—i G * * Cn •rH * •X r- cn r- <}■ 6 C"- rH o\ m <- o O 178 Figure 9-1. Effects of tyramine on blood pressure. Small box box Small pressure. blood on tyramine of Effects 9-1. Figure = tyramine, 5 ug/kg/min; middle box = 10 ug/kg/min; large box box large ug/kg/min; 10 = box middle ug/kg/min; 5 tyramine, = 1 ug/kg/min. 15 = blood pressure ( m m H g ) 160 n 80- 60- 40 -30 0 0 0 90 60 30 minutes 179 i —i ■— —i— 2 150 120 diastolic 1.4-1 1------1 1 ' 1 1 1 1 1 -30 0 30 60 90 120 150 minuted Figure 9-2. Effects of tyramine on plasma catecholamines. 180 Figure 9-3. Effects of 30 minutes cold stress on blood pressure. 181 minutes Figure 9-4. Effects of 30 minutes cold stress on plasma catecholamines. 182 Figure 9-5. Relationship between plasma noradrenaline and and noradrenaline plasma between Relationship 9-5. Figure systolic blood pressure during tyramine infusions, and cold cold and infusions, tyramine during pressure blood systolic stress. increase in systolic BP ( m m Hg) plasma noradrenalineplasma (ng/ml) 183 9.4 DISCUSSION 9.4.1 Effects of cold stress The relationship between changes in BP and in plasma noradrenaline following endogenous noradrenaline release have been examined in this study by comparing the effects of tyramine and cold stress. The technique of prolonged cold stress reported here has not been previously described as a deliberate method for stimulation of sympathetic nervous activity, even though various forms of exposure to cold are known to elevate both blood pressure and plasma noradrenaline in animals and man (LeBlanc et al, 1979; Winer et al, 1977; Johnson et al, 1977; Picotti et al, 1981). Thirty minutes exposure to cold air is in many ways a very different form of stimulus of sympathetic activity from the infusion of tyramine, but the two stimuli are clearly comparable in several important respects. Firstly, subjects are studied supine throughout, thereby avoiding postural changes with their associated haemodynamic and sympathetic effects. Secondly, the period of observation in both cases allowed sufficient time for noradrenaline concentrations to achieve a new steady state in plasma (Fitzgerald et al, 1979) - in contrast to brief stimuli such as the cold pressor test (Robertson, 1979; LeBlanc, 1979). Thirdly, tyramine and cold stress appear to be specific stimuli of noradrenaline release. Neither increases plasma adrenaline levels which might complicate the interpretation of haemodynamic 184 changes. As these conditions would not have been met by other commonly used experimental techniques for the stimulation of sympathetic nervous activity, it seemed appropriate to use the method of 30 minutes cold exposure described above. In normal subjects, cold stress induced a modest but significant elevation of both systolic and diastolic blood pressure of about 10 mmHg. Similar increases in blood pressure have been reported by other groups who have studied the haemodynamic effects of prolonged cold exposure (Hattenhauer et al, 1975), although in older subjects the rise in BP can be as much as 40 mmHg (Zenner et al, 1980). The subjects felt that the stress of prolonged cold exposure was not just a pain stimulus as experienced during the 2 minute immersion of the hand in iced water employed in the cold pressor test. The rise in systemic vascular resistance that occurs during cold exposure is likely to reflect release of nordrenaline from peripheral arterioles. However, it cannot be excluded that some of the increase may have derived from skeletal muscle which is a large contributor to circulating noradrenaline (Brown et al,1981). Plasma noradrenaline concentrations correlate well with direct recordings of muscle nerve activity (Wallin et al, 1973; Wallin et al, 1981). 9.4.2 Differential haemodynamic responses The blood pressure responses during the two procedures differed in type as well as in magnitude. During cold stress, 185 both diastolic and systolic pressures increased equally without widening of the pulse pressure; this type of response is consistent with increased peripheral vascular resistance due to arteriolar constriction, with little or no change in cardiac output. Published studies have confirmed that changes in cardiac output during exposure to cold are negligible (Epstein et al, 1969) . With tyramine, a different response was obtained. As in the previous studies diastolic pressure was unchanged, but the systolic pressure increased and the pulse pressure widened, reflecting the cardiac inotropic effects. These differing patterns of haemodynamic response (to cold and tyramine) appear to reflect the differing contributions to the heart and of the resistance vessels to the acute pressor response. Stimulation of noradrenaline release in either of these 'sectors' raises blood pressure independently and by different mechanisms; and in each case this is accompanied by an increase in circulating noradrenaline. Yet because the effects of released noradrenaline depend on both the function of the target tissue, and its contribution to the circulating noradrenaline pool, a given increment in the plasma noradrenaline level will not always be associated with the same change in blood pressure. This is evident from the dissimilar NA/SBP relationships observed for tyramine and cold stress. Clearly, even with endogenous noradrenaline release, the NA/BP relationship may vary even within an individual. 186 9.4.3 Dissociation of pressor responses from plasma noradrenaline This observation that blood pressure elevation by different pressor mechanisms may have very different effects on circulating noradrenaline levels is relevant to the concept of using plasma noradrenaline as an index of sympathetic activity, and of relating such data to haemodynamic changes. The most important implication is that pNA levels cannot be used to compare differences in sympathetic activity between individuals unless it can be assumed that their patterns of sympathetic outflow, and haemodynamic activity are similar. However, since there are many factors which may influence basal blood pressure and blood pressure responses (- for example, age, diet, autonomic function, the renin-angiotensin system, mineralocorticoid secretion, circulatory volume and vasoactive endothelial factors), it is doubtful whether the degree of similarity or otherwise can be determined with confidence in many individuals. Even within an individual, it is difficult to interpret the significance of changes in plasma noradrenaline if the sympathetic mechanisms involved are not the same. 9.4.4 Limitations of plasma noradrenaline as an index of sympathetic nervous system activity The interpretation of plasma noradrenaline levels has not always reflected the anatomical and functional complexity of the 187 sympathetic nervous system and the variable kinetics of circulating noradrenaline (Esler, 1982). There is now a growing body of experimental data which is difficult to reconcile with the original concept of generalised sympathetic activity (Goldstein et al, 1983). It has also become increasingly clear that changes in sympathetic outflow which occur during activity and arousal are often not uniform, but directed selectively at specific organs according to physiological requirements (Abboud et al, 1979; Shepherd et al, 1982). However, the amount of noradrenaline contributed by each organ to the mixed venous plasma noradrenaline (taking into account the proportion of the cardiac output received) appears to be unrelated not only to the organ's sympathetic innervation, but also to its importance in blood pressure control. Thus the heart and kidneys contribute only about 3% and 12% respectively, and the adrenals about 2% (Brown et al, 1981). Skeletal muscle, though less densely innervated, makes up some 30kg body mass in an average male (compared with 0.4kg for kidneys or heart) and is probably a dominant factor in determination of circulating noradrenaline. The splanchnic circulation and liver together receive about 25-30% of the resting cardiac output, but catecholamines in this vascular bed are efficiently cleared from venous blood by the liver (Folkow et al, 1983). In the lungs there is a modest net extraction of noradrenaline, which is important because the lungs receive the entire cardiac output. Also, the lungs also appear to be a major contributor of noradrenaline to the mixed venous pool but without having direct 188 influence on systemic pressure (Esler et al 1985). Data from a number of studies on the arterial and venous noradrenaline levels in various organs, together with the percentage net addition or extraction of noradrenaline, are summarised in table 9.3. Radiotracer studies of regional noradrenaline release in normotensive subjects show that the proportion of circulating noradrenaline contributed by individual organs were broadly in agreement with the figures obtained by estimated cardiac outputs. The results of Esler and colleagues were: lungs 30%; kidneys 25%; skeletal muscle 22%; splanchnic and hepatic (combined) 6%; skin 4%; heart 3%; adrenals 2% . Thus, the heart and kidneys, two key organs in the determination of blood pressure, together contribute only 25-30% of all noradrenaline released to the plasma. Skeletal muscle and lungs contribute over 50% (Esler et al 1985). Understanding the contribution of the major organs to circulating noradrenaline has helped to explain the apparently conflicting changes in plasma noradrnaline and haemodynamics sometimes seen with some sympathetic stimuli. The likely mechanisms are shown in table 9.4 which summarises a number of studies. These data show that with stimuli of mild to moderate intensity, the changes in sympathetic activity, noradrenaline, and haemodynamics are not necessarily consistent. 189 Table 9.3 Arteriovenous differences in plasma noradrenaline in human vascular beds. Organ NAa NAv % V-A Heart 0.297 0.380 +28% Kidneys 0.272 0.349 +28% Arms 0.275 0.338 +23% Legs 0.227 0.383 +25% Adrenals 0.223 6.412 +2,775% Liver 0.328 0.158 -52% Lungs 0.354 0.284 -20% Abbreviations: NAa, arterial noradrenaline. NAv, venous noradrenaline. V-A, venoarterial. (from Goldstein et al, 1983) Plasma noradrenaline levels therefore represent the algebraic sum of all the regional changes in noradrenaline output. It is therefore very unlikely that changes in th*£mixed venous noradrenaline pool can be used to determine regional patterns of sympathetic activity. How then should plasma noradrenaline levels be interpreted? For purely clinical purposes the use of plasma noradrenaline is well defined, since it concerns comparison of a patient's results with a normal range. For example, high levels may indicate a phaeochromocytoma; low levels are seen in 190 Table 9.4 Plasma noradrenaline and haemodynamic patterns in some sympathetic reflexes in man. stimulus HR MAP MVR MSA NA AD decrease in carotid + + (+) (+) +/- +/- baroreflex activity decrease in volume +/- +/- + ++ ++ +/- receptor activity orthostasis + +/- + ++ ++ (+) emotional stress ++ + - +/- (+) ++ Abbreviations: HR, heart rate. MAP, mean arterial pressure. CL MVR, muscle vascular resistance. MSA, muscle sympathetic activity. NA, noradrenaline. AD, adrenaline. (after Folkow et al, 1983) idiopathic orthostatic hypotension (Bravo and Tarazi, 1982). For the purposes of pathophysiological investigation, it appears that plasma noradrenaline may give a reasonable indication of sympathetic activity under highly controlled experimental conditions, as demonstrated in chapter 6. Within individuals or groups subjected to a single stimulus of at least 191 moderate intensity - pain, cold exposure or muscular activity for example, a good, even quantitative indication of the level of sympathetic activity can be inferred provided that variability in the data is minimised by experimental design. With small increases in plasma noradrenaline, the accuracy and precision of the assay method becomes increasingly important. Interpretation of the data is complicated by differences in noradrenaline handling and kinetics between the various organ beds. In normal subjects, less than 40% of circulating noradrenaline is derived from organs directly involved in maintaining blood pressure, and the effects on blood pressure and plasma noradrenaline levels may vary greatly when different mechanisms are involved. Finally in the basal state it is clear that the circulating noradrenaline level is determined equally by noradrenaline release and clearance. Therefore, when relatively small differences in sympathetic activity are concerned, the validity of peripheral venous plasma noradrenaline levels as a means of comparing sympathetic activity between individuals or groups is doubtful. 9.5 SUMMARY The significance of peripheral venous noradrenaline levels in the assessment of predominantly regional symppathetic activity was investigated by comparing the effects of tyramine and cold stress. 192 Both stimuli caused an increase in blood pressure apparently by differnt haemodynamic mechanisms, and the effects on plasma noradrenaline levels were also different. For a given rise in systolic blood pressure the increment of plasma noradrenaline during cold stress was about fivefold greater than during tyramine infusions. This demonstrates that the NA/BP relationship is not constant even within an individual. It is concluded that plasma noradrenaline may be a useful guide to short term changes of sympathetic acvtivity within a group exposed to the same conditions, but its validity for comparing small changes in sympathetic activity between groups or individuals is doubtful. Even within an individual, plasma noradrenaline levels may be of limited use when comparing stimuli which differ greatly in their patterns of regional sympathetic outflow. 193 CHAPTER 10 A METHOD FOR THE ESTIMATION OF INTRACARDIAC NORADRENALINE KINETICS 194 10.1 INTRODUCTION The limitations of peripheral venous plasma noradrenaline as an index of sympathetic nervous system activity have been considered in detail in the previous chapters. The fundemental problems can be summarised as follows. (1) It cannot be assumed that sympathetic outflow is generalised, but the relationship between peripheral venous plasma noradrenaline and regional sympathetic activity cannot be assessed. (2) The proportion of circulating noradrenaline derived from tissues involved with the control of blood pressure (heart, kidneys, adrenals) is exceeded by the proportion from tisues that are not. (3) The haemodynamic effects of an organ an its contribution to plasma noradrenaline are not related. (4) Of the noradrenaline released into the synaptic cleft, only a small proportion - about 20% - diffuses into plasma, so that any changes in sympathetic activity appear only in dampened form. (5) The noradrenaline plasma clearance rate is a major determinant of the plasma noradrenaline level, and this varies widely between individuals. In addition to this, there are a number of methodological factors (see sections 2.4.5, 2.4.6 and 3.5.2) which may which may introduce sampling and measurement errors. In view of these limitations, it would appear preferable to use techniques which are capable of assessing regional noradrenaline release as a means of studying in detail the sympathetic activation of specific organs. This necessarily 195 involves a more complex experimental methodology. Sampling the venous effluent of a particular organ would in itself be inadequate since it would depend partly on the noradrenaline concentration of the arterial inflow. Measurements based on the arteriovenous difference may be affected by changes in organblood flow. It is therefore necessary to measure arterial and venous plasma noradrenaline concentrations, as well as the plasma flow through the organ or vascular bed. 10.2 THEORETICAL BASIS FOR ESTIMATION OF INTRACARDIAC NORADRENALINE KINETICS 10.2.1 The problem of noradrenaline uptake The studies of Brown and Gillespie (1957) on the isolated perfused cat spleen first demonstrated that the release of noradrenaline by an organ is determined by the activity of its sympathetic nervous supply. This principle has been confirmed in more recent studies on the dog heart in vivo. (Yamaguchi et al, 1975; Yamaguchi and de Champlain, 1977; Levy and Blattberg, 1976a, 1976b.) Theoretically therefore, estimating the rate of norepinephrine production should provide the best index of sympathetic activity. If the sole function of sympathetically-innervated tissues was to release noradrenaline, then it would be a relatively straightforward matter in principle to determine the 196 rate of release in an organ or vascular bed. This could be achieved by obtaining the arteriovenous difference in plasma noradrenaline concentrations, and multiplying this figure by the plasma flow, to determine the total addition of noradrenaline to perfusing blood. However, as well as releasing noradrenaline (or in the case of adrenal medulla, adrenaline), sympathetically innervated tissues also continuously extract catecholamines from perfusing blood, principally by neuronal reuptake (Iversen, 1967). Because this is a continuous bidirectional process, the total noradrenaline release within an organ cannot be measured directly since the rate of simoultaneous noradrenaline uptake is unknown. Measurement of the total addition of noradrenaline to the perfusing blood ( arteriovenous NA difference x plasma flow) gives only the net result of these two opposing processes, namely uptake and release. (This figure, the "apparent" release, will be termed the overflow.) Clearly, in order to determine the cardiac release of noradrenaline, it is necessary first to determine its rate of uptake; this, when added to overflow, will provide an estimate of the total intracardiac release of noradrenaline. To summarise: NA overflow = intracardiac NA release - NA uptake Thus, Intracardiac NA release = NA overflow + NA uptake 197 10.2.2 Terminology Intracardiac noradrenaline release. This refers to the noradrenaline which, having been released from the nerve terminal, is not removed by the uptake processes but spills over into the extracellular fluid and bloodstream. It reflects the true rate of release which is probably several fold greater. (The term 'release' is preferred to 'spillover' in order to avoid confusion with 'overflow'.) The units are ng/min . Uptake. This refers to the extraction of adrenaline and noradrenaline by cardiac tissues. It is a result of neuronal uptake, uptake-2, and metabolism. The units are ng/min. Overflow. This refers to the net addition of noradrenaline to the blood perfusing the heart. It is directly measured as arteriovenous noradrenaline difference x plasma flow, and given in units of ng/min. 10.2.3 Measurement of noradrenaline overflow This is given by direct measurements. NA overflow = (NAao - NAcs) x PF where: NAao = aortic plasma NA concentration (ng/ml) NAcs = coronary sinus NA concentration (ng/ml) PF = plasma flow (ml/min) = coronary blood flow x (1-haematocrit) 198 10.2.4 Estimation of noradrenaline uptake At present, direct measurement of cardiac noradrenaline uptake in man is technically difficult. It can be achieved by radiotracer and pharmacological methods (Esler et al, 1983; Goldstein et al, 1988) but these methods were too complex for the present study. In the present study, estimates of noradrenaline uptake were derived from direct measurements of adrenaline uptake. This is because adrenaline (unlike noradrenaline) is synthesised and released only by the adrenal medulla: the fractional uptake of adrenaline by the heart can therefore be obtained by measuring the adrenaline arteriovenous difference. Noradrenaline uptake is then estimated by assuming that the percentage uptake of noradrenaline entering the heart (ie NAao x PF) is the same as that for adrenaline. ie NA uptake (ng/min) = fractional AD uptake x NAao x PF = ADao - ADcs x NAao x PF ADao The validity of this assumption depends on two observations. First, kinetic data from in vitro studies (mainly on rat heart) show that the two amines share the same uptake 199 mechanisms, and have very similar affinities for the uptake process (Iversen, 1967; Gillespie, 1973.) Second, data obtained during selective venous sampling in patients with suspected phaeochromocytomas demonstrated that the percentage extraction of adrenaline and noradrenaline in vivo was similar through a wide range of plasma concentrations, and in a number of organs including the heart (Brown et al, 1981) 10.2.5 Estimation of intracardiac noradrenaline release: sample calculation. This is given by NA release (ng/min) = NA overflow + NA uptake. Assuming: aortic noradrenaline = 0.30 ng/ml coronary sinus noradrenaline = 0.50 ng/ml aortic adrenaline = 0.08 ng/ml coronary sinus adrenaline = 0.04 ng/ml coronary blood flow = 120 ml/min haematocrit = 0.40 1-haematocrit = 0.60 NA overflow = (0.50 - 0.30) x 120 x 0.60 = 14.4 ng/min 200 NA uptake 0.08 - 0.04 x 0.30 x 120 x 0.60 0.08 = 10.8 ng/ml NA release = 14.4 + 10.8 = 25.2 ng/min 10.3 DISCUSSION This method of estimating intracardiac noradrenaline kinetics is attractive for its conceptual simplicity, but can be justly criticised for several reasons. First, selective arterial and cardiac catheterisation is required. These invasive techniques can only be justified in patients with suspected cardiac disease in whom diagnostic catheterisation is clinically indicated. It is recognised that data obtained in this selected group of subjects may be misleading. Second, the calculated estimates are highly dependent on accurate assay of both catecholamines, but especially adrenaline. The nature of the calculations ensures that any error will be greatly compounded by successive multiplications. Inevitably, this will tend to increase the variability of the data such that statistical confirmation of apparent differences will be more difficult to obtain. Third, the present method has not been validated against 201 any other method. Alternative methods which have been described for assessment of cardiac uptake and/or release of nordrenaline include radiotracer techniques and pharmacological methods based on inhibition of uptake (Goldstein et al, 1988). The facilities and expertise required for these methods were not available in our institution. However, the results obtained during this study are supported by data subsequently published by Goldstein et al (1988). These investigators used radiolabelled noradrenaline to estimate cardiac noradrenaline uptake, although in other respects their calculations were identical to those described above. They obtained a figure of 12.9 ng/min for resting cardiac noradrenaline release, compared with 25.5 ng/ml in the present study. These results may be closer than at first appears. It should be noted that our patients had a (directly measured) step-up from arterial to venous noradrenaline Of 0.421 to 0.611 ng/ml, whereas the arterial and venous levels in Goldstein's group similar, viz 0.210 and 0.217 ng/ml. As our patients were net 'producers' of noradrenaline, their rate of intracardiac noradrenaline release may well have been genuinely higher, particularly since uptake was similar in both studies: 11.9 ng/ml by Goldstein, and 10.1 ng/min is this study. Therefore, if our patients had also had equal arterial and venous noradrenline levels (ie uptake = release), then their calculated noradrenaline release would have been 10.1 ng/min, very similar to the 12.9 ng/min reported by Goldstein et al (1988) . 202 CHAPTER 11 CARDIAC NORADRENALINE RELEASE AT REST AND DURING COUPLED RIGHT VENTRICULAR PACING 203 11.1 INTRODUCTION The purpose of this study was to assess intracardiac noradrenaline kinetics at rest and during a period of physiological sympathetic activation that could be sustained for 5-10 minutes (to allow equilibration of noradrenaline levels) without discomfort or risk to the patient. The administration of tyramine was considered ethically inappropriate since data concerning its safety in cardiac disease are lacking. Previous experience as well as published data on the technique of 'coupled' right ventricular pacing suggested that this might be a stimulus associated with sustained sympathetic activation of the heart (Geschwind, unpublished observations; see also references cited below). This requires positioning of a pacing electrode in the apex of the right ventricle. A programmable external pacemaker generator is set to deliver a single extrastimulus just after recovery of ventricular refractoriness following each successive sinus beat ie as early as possible. This produces a sustained ventricular bigeminal rhythm. In this bigeminal rhythm, the (paced) extrastimuli are so premature that their stroke output is negligible, whereas all the sinus beats show maximal compensatory enhancement of ventricular function, ie post-extrasystolic potentiation. This effect has been used to study left ventricular contractile reserve and identify segmental wall motion 204 abnormalities in ischaemic heart disease (Cohn, 1980) . The degree of enhancement of the postextrasystolic contraction increases with the prematurity of the extrasystole (Palmero 1976), but the fundamental mechanism of the potentiation is still uncertain. Current evidence (Suko et al, 1970) suggests that the enhanced myocardial contractility may be related to increased availability of intracellular calcium at the contractile sites. The changes in preload and afterload that occur during the compensatory pause are now regarded as having a contributory rather than a major role in the enhancement of contractile state (Sung et al, 1980). Since catecholamines are known to increase calcium influx by alteration of the cellular gating mechanism (Lindemann, 1982), the present study was undertaken to determine whether catecholamines have a role in the enhancement of ventricular function during postextrasystolic potentiation. 11.2 AIMS OF THE STUDY The aims of this study were to assess intracardiac noradrenaline kinetics at rest and during coupled right ventricular pacing. Also to determine whether there was evidence in favour of sustained sympathetic activation during postextrasystolic potentiation of cardiac contractility. 205 11.3 METHODS 11.3.1 Patients A total of thirty patients with mild or no symptoms, haemodynamically insignificant cardiac abnormalities, and normal left ventricular function were studied during routine cardiac catheterization (details are given in table 11.1). Five took medication, but this was discontinued at least 24 hours before the study. All studies were performed in the morning with the patient in the fasting state, supine and premedicated with lOmg of diazepam intramuscularly. Before the experimental study, all patients underwent diagnostic cardiac catheterisation that included biplane left ventricular cineangiography and coronary arteriography. Patients with coronary artery stenoses of greater than 50%, or with a left ventricular ejection fraction of less than 0.45 were excluded from the study. For ethical reasons, it was decided not to obtain haemodynamic, angiographic, metabolic, and catecholamine data simultaneously in all patients since a long and unjustifiably arduous protocol would be required. Thus, although all patients had the identical pacing protocol, angiographic and ventricular function data were obtained in 15 patients (Group 1), while in the remaining 15 patients (group 2), cardiac catecholamines and peripheral haemodynamic data were recorded. In other words, the 206 two groups were studied 'in parallel'. This protocol was approved by the Research Ethics Committee of CHU Henri Mondor. Informed consent was obtained from all patients. As shown in table 11.1, the composition of both groups in terms of age, sex, diagnosis and functional class was similar; biplane left ventricular ejection fraction at rest was 58+2% in Group 1 and 56+2% in Group 2. 11.3.2 Protocol After completion of the diagnostic catheter study (using a percutaneous Seldinger technique from the femoral vessels), and the positioning of additional catheters, patients from both groups rested undisturbed for 10 minutes. Control data were then recorded. Following this coupled right ventricular pacing was started, and continued for 10 minutes. The data were measured again toward the end of this 10 minute period, while pacing was in progress. The following data were obtained in the two groups, at rest and during coupled pacing. Group 1. heart rate and arterial blood pressure left ventricular systolic and end-diastolic pressure peak first derivative of left ventricular pressure (dP/dt) left ventricular ejection fraction mean velocity of circumferential fibre shortening 207 Group 2. heart rate and arterial blood presure coronary blood flow simoultaneous aortic and coronary sinus blood samples for catecholamines/ haematocrit and myocardial oxygen consumption 11.3.3 Haemodynamic studies In Group 1, left ventricular pressure was continuously monitored with a 7F high fidelity microtransducer-tipped catheter equipped with a side hole for simultaneous aortic pressure recording (Millar Instruments). These data were recorded on a Bell and Howell multichannel recorder and stored in Syscomoram computer system (SNIAS, Bordeaux) for processing and retrieval. A similar arrangement was used for studies in Group 2/ but the catheter was a 7F pigtail (USCI) positioned in the ascending aorta for blood pressure monitoring. Left ventricular volumes were computed from biplane cineangiograms using a computer-based calculation for Simpson's rule. The ejection phase indices (ejection fraction and mean velocity of circumferential fibre shortening) were automatically calculated from these values. In Group 2, coronary blood flow was measured in triplicate by a continuous infusion thermodilution technique 208 using a constant infusion rate of 60ml of 0.9% sodium chloride per minute for 30 seconds (Ganz et al, 1971). This was performed using a two thermistor catheter (Wilton Webster) which was inserted through the femoral vein and advanced to the coronary sinus. The position of the sensing thermistor was confirmed fluroscopically by small injections of contrast medium. 11.3.4 Coupled right ventricular pacing Continuous postextrasystolic potentiation was induced in both groups by sustained coupled right ventricular pacing using a bipolar electrode catheter (USCI) which was advanced from the femoral vein to the apex of the right ventricle. Pacing extrastimuli were generated by a 5837 Medtronic pacemaker. Single electrical stimuli triggered by the spontaneous QRS complex and adjustably delayed were applied to the endocardial surface of the right ventricle. The extrastimulus was moved progressively earlier in the cardiac cycle to achieve maximal prematurity. This was generally 280 to 310 ms after the preceding R wave before encountering ventricular refractoriness. In this way, an insignificant left ventricular pressure response to the stimulus was achieved while the effect of the potentiated beat was maximised. This was was reflected by the increase in dP/dt without any significant change in peak systolic left ventricular pressure. The pressure pulse generated by the extrastimulus was negligible and usually seen 209 as a notch on the descending limb of the left ventricular pressure curve. Since the contractile force of the extrasystole was insufficient to open the aortic valve (and cause a rise in aortic pressure), the aortic pressure pulses were caused by the potentiated beats alone. Coupled pacing with the fixed interval was maintained for 10 minutes to secure a steady state of potentiation. 11.3.5 Biochemical studies In Group 2, blood samples were taken for catecholamines and myocardial oxygen consumption. Oxygen partial pressures were determined by a polarographic method using a BMS 3 radiometer apparatus. Oxygen binding capacity was measured using a LEX-02-CON oximeter (Lexington Instruments). Oxygen content was calculated from oxygen partial pressures, oxyhaemoglobin binding capacity and haemoglobin. Blood samples for catecholamines were handled, stored and assayed as previously described. 11.3.5 Intracardiac noradrenaline kinetics This method was described in the preceding chapter. 210 11.4 RESULTS 11.4.1 Heart rate and blood pressure In both groups, coupled pacing caused a profound decrease in the (effective) arterial pulse rate because only the potentiated (and not the extrasystolic) contractions generated arterial pressure pulses (figure 11-1). Thus, the control heart rate in Group 1 was 77+7 bpm and this fell to 43+4 bpm (p<0.001); the fall in the arterial pulse rate was similar in Group 2, from 73+4 to 42+3 bpm (p<0.001). The fall in the arterial pulse rate was accompanied by an immediate fall in the diastolic pressure, chiefly as a consequence of the lengthened interval between beats. The systolic pressure was unchanged. Thus, the mean aortic blood pressure fell, within the duration of a single cardiac cycle, from 85+15 to 74 +11 mmHg in Group 1, and from 91+10 to 7 94^10 mmHg in Group 2 (p<0.001 for both). Thus, in both groups, the decrease in mean aortic pressure was similar and due principally to a decrease in the diastolic component because systolic pressure did not change during coupled pacing. 11.4.2 Group 1: Effects of coupled pacing on left ventricular function In group 1, the ejection fraction, mean rate of 211 circumferential fibre shortening and left ventricular peak dP/dt (figure 11-2) all increased significantly during coupled pacing (p<0.001). This response was observed for all three variables of left ventricular function, and in all patients (tables 11.3A and 11.3B). 11.4.3 Group 2: Coronary blood flow and myocardial oxygen consumption In Group 2, coronary blood flow increased in all patients during coupled pacing from 142+43 to 1774^60 ml/min (p<0.001; see figure 11-3). This was associated with an increase in myocardial oxygen consumption, from 16+6 to 21+10 (p<0.001). 11.4.4 Plasma noradrenaline and adrenaline The resting aortic and coronary sinus plasma noradrenaline levels were 0.421 and 0.611 ng/ml respectively, showing a net release of noradrenaline by the heart. During coupled pacing, these levels increased in all 15 patients to 0.576 and 0.836 ng/ml, respectively (figure 11-4 and 11-5). Plasma adrenaline levels in aortic and coronary sinus blood were 0.083 and 0.054 ng/ml respectively showing a net extraction of adrenaline in the coronary vascular bed. These levels did not change significantly during coupled pacing and the arteriovenous adrenaline difference was also unchanged (tables 11.4A and 11.4B). 212 11.4.5 Overflow, uptake and release of noradrenaline in the heart The mean arteriovenous noradrenaline difference increased slightly, from 0.190ng/ml during basal conditions to 0.260 ng/ml during coupled pacing. However, because individual responses were more variable (11 subjects with increased and 4 with reduced AV difference during coupled pacing), this increase was not significant. For the same reason, although the overflow of noradrenaline from the heart increased by 68%, from 15.1+2.5 to 25.5+4.5 ng/min, this change just failed to reach significance (p=0.056 in the two tailed t test; figure 11-6). The basal rate of adrenaline uptake was 2.23+0.41 ng/min (35+1%), and this remained unchanged during coupled pacing at 2.41+0.55 ng/min (26+5%). The estimated noradrenaline uptake based on these figures were 10.46+6.7 ng/min at rest, and 14.22+11.8 during pacing (figure 11-7). The calculated rate of noradrenaline release within the heart was 25.6+2.8ng/min at rest; this increased significantly during coupled pacing to 39.7+5.2 ng/min (p<0.05) . See figure 11-8 and table 11.5. None of the patients experienced subjective discomfort during the study protocol, nor were any changes in their clinical condition noted. 213 Table 11.1 Group 1. Cinical details Pt age sex diagnosis NYHA LVEF 1 42 F MS I 70% 2 36 M CAD II 61% 3 18 M ASD I 66% 4 24 M ASD I 52% 5 35 M MS I 56% 6 34 M MS II 52% 7 53 F MS I 56% 8 41 F ASD I 58% 9 45 M CAD I 64% 10 42 M MS I 70% 11 40 M AR I 73% 12 41 M AR II 45% 13 55 F AR II 54% 14 39 M AR II 51% 15 35 M AR II 51% Abbrevations. AR, aortic regurgitation. ASD, atrial septal defect. CAD, coronary artery disease. MS, mitral stenosis. MVP, mitral valve prolapse. NYHA, New York Heart Association functional class. LVEF, left ventricular ejection fraction. 214 Table 11.2 Group 2. Clinical details Pt aqe sex diagnosis NYHA LVEF 16* 41 M CAD I 48% 17* 30 M AR II 58% 18* 40 F MVP II 64% 19* 42 M AR II 54% 20* 48 M PDA I 55% 21* 33 M AR II 49% 22* 39 F MS II 49% 23 56 M AR II 55% 24 36 F MS I 65% 25 45 M CAD II 45% 26 31 M ASD I 65% 27 39 F MS I 60% 28 50 M MS I 53% 29 47 F MS II 54% 30 49 F MS I 69% Abbreviations as in table 11.1 * patients in propranolol study (see chapter 12). 215 Table 11.3a Group 1. Haemodynamics at rest. Pt pulse MAP LV dP/dt LVEF VcF 1 85 100 1700 70% 1.6 2 88 92 1800 61% 1.3 3 76 89 1500 66% 1.6 4 83 95 1500 52% 1.1 5 76 80 1400 56% 1.4 6 70 74 1800 52% 1.3 7 67 74 1700 56% 1.1 8 75 105 1900 58% 1.5 9 80 93 1700 64% 1.5 10 94 93 1700 70% 1.6 11 72 82 1400 73% 1.6 12 71 71 900 45% 1.0 13 69 70 1000 54% 1.1 14 75 81 1200 51% 1.4 15 76 79 1000 51% 1.0 mean 77 85 1480 59% 1.3 sd 8 15 320 7% 0.3 Abbreviations and units. Pulse (bpm) . MAP, mean arterial pressure (mmHg). LV dP/dt, peak first derivative of left ventricular pressure (mmHg.sec). VcF, mean velocity of circumferential fibre shortening (circumferences/sec). 216 Table 11.3b Group 1. Haemodynamics during coupled pacing. Pt pulse MAP LV dP/dt LVEF VcF 1 43 70 3000 79% 2.1 2 45 82 3800 77% 2.0 3 40 71 3000 72% 2.1 4 41 92 3000 64% 1.5 5 49 73 3100 72% 1.8 6 38 61 2600 76% 2.1 7 42 57 2700 83% 2.5 8 43 98 4200 77% 1.8 9 44 79 2100 79% 1.9 10 52 83 2600 74% 2.1 11 45 73 2000 77% 2.0 12 34 61 1550 71% 1.5 13 39 65 1300 60% 1.7 14 44 71 1550 66% 1.7 15 43 70 1500 72% 1.1 mean 43* 74* 2530* 73%* 1.8* sd 4 11 850 8% 0.4 Abbreviations etc. as in table 11.3a. *p<0.001 compared with rest. 217 OCMOi—I CM O^ i—Ih'NvDvXJONOst'CO in Mr-4 r-ir^-r^ cd 0 0 vo VO in rH CO CO m rH CM VO oo o o rH rH M i-H VO o VO 0 0 CO Oi oo vo 0 0 CO CM CJ\ r>- rH OO o o 6 m in ON OO < r in oo m m oo vo rH X < \ o bO o o o o o o o o o o o o o o o O o cd vo OO CM o\ ON oo vo CO m m VO vo rH ME o rH OO r->»vo r- 1—1 CO Ov m oo vo vo oo CM 1-1 cd cd B CO m OO CO CO m CM CM CO OO m uo 1-1 rH \ 218 O CO CMo VO O' rH r- CMO' 00 O oo VO rH corH CMrH m VO O' m 00 in m n- 1"- CO rH CO vO CO O s rH rH o o o O o O o o o o o o o o o 00 Q \ £ GO £ CO CMm o o m CMVO oom CO EH Pn rH rH rH rH CMCMCMCMCMCMCMCM CM1 CM1 CO cd co Abbreviations as in table 11.4a. **p<0.01 compared with rest cu B 219 Table 11.5 Cardiac catecholamine turnover. Mean+SD REST PACING AD uptake 2.2+1.5 2.4+2.1 (ng/min) AD % uptake 34.9+25.5 26.8+20.0 (%) NA uptake 10.5+6.7 14.2+11.8 (ng/min) NA overflow 15.1+9.6 25.4+17.4 (ng/min) NA release 25.5+10.0 39.7+20.3* *p<0.05 220 COUPLED PACING 1401- Figure 11-1. Simoultaneous recording of the electrocardiogram (ECG, upper trace) and the aortic blood pressure (AP, lower trace) at the onset of coupled pacing (arrowed). The ECG shows ventricular paced beats coupled to sinus beats. The pressure trace shows the immediate slowing of the (effective) pulse rate, and the fall in diastolic and mean (M) blood pressure. 221 LV dP/dT Figure 11-2. Group 1. The effects of coupled pacing on the peak first derivative of left ventricular pressure. 222 coronary blood flow 400 300 - c E 200 - E 100 - 0 REST COUPLED PACING Figure 11-3. Group 2. Increased coronary blood flow (CBF) during coupled pacing. 223 aortic plasma NA Figure 11-4 Group 2. Aortic plasma noradrenaline levels at rest and during coupled pacing. 224 coronary sinus plasma NA Figure 11-5. Coronary sinus plasma noradrenaline levels at rest and during coupled pacing. 225 cardiac NA overflow Figure 11-6. Group 2. The net overflow of noradrenaline from the heart at rest and during coupled pacing. v 226 cardiac NA uptake Figure 11-7. Group 2. Noradrenaline uptake by the heart at rest and during coupled pacing. 227 intracardiac NA release 100 Figure 11-8. Group 2. Intracardiac noradrenaline release at rest and during coupled pacing. 228 11.5 DISCUSSION 11.5.1 Mechanical Effect of Postextrasystolic Potentiation. The phenomenon of postextrasystolic potentiation of cardiac contractility appears to be a fundamental property of mammalian myocardium. It was first described by Langendorff in 1885, but the mechanism of the inotropic effect is still not fully understood. There is evidence that three types of mechanism may contribute to this effect. First, in vitro preparations demonstrate that postextrasystolic potentiation is associated with increased availability of intracellular calcium which occurs regardless of changes in preload and afterload (Braunwald et al, 1964; Hoffman et al, 1965; Ross et al, 1965). In addition, studies in intact animals (Lendrum et al, 1960) suggest that the fall in aortic impedance during the compensatory pause may also enhance contractility, although the magnitude of this effect is uncertain. Finally, Starling's law predicts that increased ventricular filling during the compensatory pause may also contribute to po^stextrasystolic potentiation, but a recent study (Sung et al, 1980) showed enhanced left ventricular function in the absence of increased left ventricular end-diastolic volume. The results in Group 1 show that sustained coupled pacing greatly enhanced left ventricular function in all patients. The 229 increase in left ventricular peak dP/dt, an isovolumic index relatively independent of afterload (Van Den Bos et al, 1973), suggests that this was due to an improvement in the inotropic state. The increase in contractility was remarkably uniform among all patients regardless of the type of cardiac disease, and this agrees well with findings from previous reports (Cohn, 1980; Sung et al, 1980; Schwarz et al, 1975; Hirshfield et al, 1974; Popio et al, 1977; and Nesto et al, 1982)). Taking into account the similarity of left ventricular function between the two groups as judged by ejection fraction and clinical state, it is reasonable to assume that a similar increase in contractility occurred in the patients in Group 2. 11.5.2 Evidence for increased cardiac sympathetic activity The role of endogenous catecholamines in postextrasystolic potentiation has not been previously investigated. This study demonstrates that sustained coupled pacing is associated with an increased rate of noradrenaline release in the myocardium, a striking increase in cardiac contractility, and an increase in myocardial oxygen consumption. These data strongly suggest that coupled pacing caused an increase in cardiac sympathetic activity. The enhancement of sympathetic activity within the heart is also supported by the marked increase in myocardial oxygen consumption during coupled pacing, which is in agreement with the findings reported by others (Huet et al, 1981 and Simonsen 230 et al, 1978) . This occurred despite substantial reduction in the effective heart rate, possibly as a result of the energy demands associated with the inotropic effect of sympathetic stimulation. This would be in keeping with the observation of Boerth et al (1978) that for a given increase in inotropic state, myocardial energy requirements were greater when the increase was caused by noradrenaline than when it was caused by increased heart rate. Some of the increased noradrenaline release might have occurred as a conequence of pacing-induced depolarisation. However, in this study, the total number of depolarisations (ie sinus + paced) increased by only 11%, and this is unlikely to have been the sole cause of the increase in noradrenaline release. 11.5.3 Mechanism of sympathetic activation A number of factors suggest that the increase in cardiac noradrenaline release may have resulted from stimulation of a sympathetic reflex. There was a small but significant increase in (arterial) plasma noradrenaline during coupled pacing, and it is most unlikely that it could have resulted from cardiac spillover alone. Nor can it be explained by the action of pacing per se, since cardiac pacing alone does not stimulate peripheral sympathetic activity (Schwartz et al, 1979) On the other hand, since coupled pacing brought about a 231 abrupt and substantial fall in blood pressure, it is likely that some activation of high-pressure carotid and aortic arch baroreceptors would have occurred. This would have resulted in a reflex increase in sympathetic activity which would have led to an increase in cardiac and peripheral noradrenaline release. Some of the reflex cardiac stimulation could also be mediated via the parasympathetic system, leading to vagal disinhibition. Removal of the inhibitory vagal presynaptic tone can enhance the release of noradrenaline (Levy and Blattberg, 1976). This arm of the reflex appears to capable of the very rapid response times necessary for beat to beat modulation of cardiac function (Kircheim, 1976). 11.5.4 Cardiac and peripheral sympathetic activity This study confirms that assessment of regional sympathetic activity can provide considerable insight into cardiovascular mechanisms. Arguably, this also enhances the value of peripheral plasma noradrenaline in some cases since its origin can be more reliably assessed. It may also be easier to evaluate its significance if there is also information available about cardiac sympathetic activity. 232 CHAPTER 12 THE EFFECTS OF PROPRANOLOL ON CARDIAC NORADRENALINE RELEASE 233 12.1 INTRODUCTION The results reported in chapter 11 show that the method described for estimation of cardiac noradrenaline kinetics can detect increased intracardiac noradrenaline release. This was apparently consistent with increased cardiac sympathetic activity as judged by metabolic and haemodynamic data. The purpose of the present study was to assess the effects of propranolol on cardiac and peripheral sympathetic activity, since this compound is usually associated with reduction in sympathetic responses. Despite the widespread use of propranolol and the considerable efforts directed to inve^igating its effects in hypertension, angina and myocardial infarction, there is still little agreement about its effects on sympathetic nervous activity both in the heart and in the peripheral circulation. In animal studies it has been possible to assess sympathetic neural activity by direct recordings from peripheral nerves; most though not all of these studies have reported that propranolol reduced sympathetic nerve firing (Chevalier-Cholat et al, 1978; Tuttle and McCleary, 1978; Korner, 1980; Lewis and Haeusler, 1975.). In man, however, the use of plasma noradrenaline levels to assess changes in sympathetic activity has given conflicting results. Some groups have observed small increases in resting plasma noradrenaline after acute intravenous (Hausen et al, 234 1987; Galbo et al, 1976) and oral (Rahn et al, 1978) proranolol findings which appeared consistent with the associated elevation of peripheral vascular resistance (Ulrych et al,1969; Tarazi et al, 1972; Hansen et al, 1974) This led to the suggestion that propranolol stimulates a reflex increase in peripheral sympathetic outflow secondary to a reduction in cardiac output (Ulrych et al, 1969; Tarazi et al, 1972). Equally, however, comparable studies with propranolol have shown unchanged (Bonelli et al, 1979; Maling et al, 1979; van Baak et al, 1982) or even reduced (Vitanen et al, 1979) plasma noradrenaline levels, consistent with the opposite view that propranolol attenuates sympathetic outflow as shown in animal studies. Data concerning the effects of propranolol on noradrenaline release in the heart are lacking. 12.2 AIMS OF THE STUDY The aims of this study were to assess the effects of propranolol and of sympathetic stimulation, seperately and in combination, on cardiac as well as peripheral noradrenaline release. As in the previous study, the technique of coupled pacing was used as the sympathetic stimulus. 12.3 METHODS 12.3.1 Patients 235 We studied 8 patients with mild or absent symptoms (NYHA Class I-II) referred for evaluation of valvular, coronary artery or congenital heart disease, whose lesions proved to be haemodynamically insignificant. Details are given in table 11-1. None of the patients had previously had heart failure; all had angiographically normal left ventricular function. One patient had clinically insignificant coronary artery disease (stenoses < 50%) without previous myocardial infarction or symptoms of angina pectoris. Only two patients had regular medication (one took a diuretic, and the other a long-acting nitrate) but this was witheld for 24 hours before the study. All patients gave informed consent, and the protocol was approved by the Research Ethics Committee of CHU Henri Mondor. 12.3.2 Protocol As previously described, additional cardiac catheters were placed after the end of the diagnostic study, and the patients rested for 10 minutes. The study took place in 4 parts as listed below; each part lasted for 10 minutes, and the patients rested for 5 minutes between parts. (1) control: supine rest (2) continuous coupled right ventricular pacing (3) intravenous propranolol O.lmg/kg at rest (no pacing) (4) continuous coupled right ventricular pacing after propranolol 236 Data were recorded at 7 -10 minutes of each part. These were (a) aortic blood pressure and pulse rate (b) coronary sinus blood flow and (c) simoultaneous aortic and coronary sinus blood samples for estimation of catecholamines and oxygen saturation. This was usually accomplished in 3 - 5 minutes. The materials and experimental methods were as previously described. 12.4 RESULTS 12.4.1 Heart rate and blood pressure Figure 11-1 shows the transition from resting sinus rhythm to coupled pacing. This produced extrasystoles and sinus beats (electrocardiogram, upper trace). Simoultaneous recording of aortic pressure (lower trace) shows that only the sinus beats produced aortic pressure pulses. Haemodynamic effects are summarised in table 12.2. The onset of coupled pacing was associated with an immediate and abrupt fall in the aortic pulse rate from 75+10 to 42+5 bpm. Diastolic blood pressure also fell (because of the longer pulse intervals) from 81+9 to 65+10 mmHg. There was a small but statistically insignificant rise in systolic blood pressure. As soon as the pacing was switched off, the heart rate and blood pressure returned to their baseline values within 1-2 minutes. 237 The effects of propranolol at rest were to slow the resting pulse rate to 63+8 bpm, and to lower slightly the systolic pressure to 129+15 mmHg. However, the haemodynamic response to coupled pacing was not altered: the pulse rate fell from 63+8 to 39+3 bpm; the diastolic pressure fell from 80+11 to 60+14 mmHg. The systolic pressure was again unchanged. Myocardial oxygen consumption and coronary blood flow were both increased during coupled pacing, but this was abolished by propranolol (table 11.2). Under all conditions coronary blood flow was closely correlated with myocardial oxygen consumption (r=0.95 to 0.98) and the slope of the CBF/MV02 relationship was the same irrespective of intervention. 12.4.2 Cardiac noradrenaline release and peripheral venous plasma noradrenaline Individual data are shown in tables 12.3 - 12.6. The mean results have been summarised in table 12.7. This demonstrates the variability of the raw uptake and release data, and emphasises the importance of sensitivity and precision in the catecholamine assay method. Coupled pacing alone was associated with a rise in plasma noradrenaline from 0.404+0.09 to 0.556+0.17 ng/ml. The (calculated) rate of noradrenaline release in the heart also rose, from 18.8+11.2 to 43.1+26.9 ng/min. There was a similar 238 increase in the measured overflow of noradrenaline in coronary sinus blood, from 11.8+8.0 to 31.1+23.1 ng/min. Myocardial oxygen consumption and coronary blood flow also rose. After the administration of propranolol the plasma noradrenaline level of 0.431+0.11 was unchanged compared with the resting level of 0.404+0.09 ng/ml. By contrast, the (calculated) cardiac noradrenaline release was elevated at 42.0+14.2 ng/min; this was associated with an elevated noradrenaline overflow of 26.3+7.3 ng/min. Propranolol modified the effects of coupled pacing. During coupled pacing (in the presence of propranolol) there was no further change in plasma noradrenaline, nor in cardiac noradrenaline release or overflow. Also, the previous pacing-induced increases in myocardial oxygen consumption and coronary blood were abolished (table 12.7). 12.4.3 Adrenaline uptake The rate of adrenaline uptake by the heart at rest varied widely between individuals, as shown in tables 12.3 to 12.6. Both total and percentage adrenaline uptake increased significantly after propranolol administration but not at any other time. Noradrenaline uptake data are given in the tables but are not considered further since they are derived from adrenaline uptake. 239 Table 12.1 Patients' clinical details. case age sex diagnosis NYHA LVEF 1 -Hct 1 42 m aortic regurgitation II 54% 0.58 2 48 m patent ductus arteriosus I 55% 0.60 3 33 m aortic regurgitation II 49% 0.54 4 39 f mitral stenosis II 49% 0.57 5 41 m coronary artery disease I 48% 0.60 6 30 m aortic regurgitation II 58% 0.58 7 40 f mitral valve prolapse II 64% 0.60 8 56 m coronary artery disease I 76% 0.57 Abbreviations: NYHA, New York Heart Association functional class. LVEF, left ventricular ejection fraction. Hct, haematocrit. 240 Table 12.2 Haemodynamic effects of propranolol and coupled pacing. Control Propranolol rest pace rest pace systolic pressure 136+16 143+19 129+15 128+17 (mmHg) X diastolic pressure 81+9 65+10 80+11 60+14 (mmHg) ++■ aortic pulse rate 75+10 42+5 63+8 39+3 (bpm) XX coronary blood 148+54 197+73 139+59 148+53 flow (ml/min) myocardial 02 17.3+2.8 24.6+4.0* 16.7+2.7 18.2+2.5 consumption (ml/min) *p<0.05, **p<0.01, ***p<0.001 compared with rest (control). +p<0.05, ++p< 0.01, +++p<0.001 compared with rest (propranolol) 241 > »H m i"- ON o oo 00 O B • o Ss on o 242 m G > •rl <}■ CO m rH m m m rH i—1 o 6 5: <3 \ VO o o in 00 o CO rH CO o oo CM 00 CO rH rH CO 243 4-1 0 > •H CO vo oo00 CO CO rH CO CO O e £ <5 \ CMCM 00 CO VO o !Z 60 CMCMco CMCMI—1 CO CM CM rH 0 • CH w P 0 Q) /-s •H0 >O 0 co r—1 •H LO 244 4h o ^ *rl o> o tH 00 CM on r-« in o s <3 \ ro o vo tH I''- rH o m vo z oo rH m tH CM CM CO CM tH rH c o rH o c id 0 M iH *H o O in vo ooCM vo M 0 /*\ VO VO CM On r-* O rH OO o VO VO O rH VO CM 1"- VO tH t"'- CO VO ON ON • cd S m CM m CO CM VO r-x in iH CM <3 \ rH Z 00 o O o O O O o o o o e r—I0) ^ .n id 4J CBF coronary blood flow. Up, uptake. %Up, percent uptake. Rl, release. Ovf, overflow. E-i Ph Hcsro 245 Table 12.7 Mean (+SD) catecholamine results Control Propranolol rest pace rest pace NA aortic 0.404 0.556 * 0.431 0.490 (ng/ml) 0.090 0.167 0.115 0.196 NA coronary sinus 0.558 0.842* 0.818** 0.819 (ng/ml) 0.161 0.219 0.272 0.296 AD aortic 0.064 0.098* 0.141 0.079 (ng/ml) 0.015 0.034 0.101 0.064 AD coronary sinus 0.049 0.086* 0.075 0.056 (ng/ml) 0.023 0.026 0.063 0.047 Coronary blood 148 197 ** 139 148 flow (ml/min) 54 73 59 53 AD % uptake 28.2 19.0 42.6* 29.9 (%) 26.4 14.1 23.8 27.3 NA uptake 7.0 11.9 15.7 11.5 (ng/min) 7.1 9.0 11.7 12.6 NA release 18.8 43.1 * 42.0** 37.3 (ng/min) 11.2 26.9 14.2 16.6 NA overflow 11.8 31.5 * 26.3** 25.7 (ng/min) 8.0 23.1 7.3 16.5 *p<0.05, **p<0.01 , ***p<0.001 compared with rest (control) +p<0.05, ++p<0.01 , +++p<0.001 compared with rest (propranolol) 246 12.5 DISCUSSION 12.5.1 Effects of propranolol at rest In this study the effects of propranolol on cardiac and peripheral sympathetic activity were assessed by estimating the uptake and release of catecholamines within the heart, and by measuring noradrenaline in the plasma. The results show that acute, intravenous propranolol administration appeared to have different effects on resting cardiac and peripheral sympathetic activity. The plasma noradrenaline level was unchanged, while by contrast, the rate of cardiac noradrenaline release increased more than twofold. This calculated figure was supported by a parallel increase in the directly-measured rate of noradrenaline overflow. These results suggest that, while peripheral sympathetic activity was unchanged, the administration of propranolol led to a marked increase in sympathetic outflow to the heart. The mechanism of this response is uncertain. But, in view of the negatively inotropic effects of acute beta blockade, it would be consistent with a reflex increase in sympathetic activity in response to reduced in ventricular function. The beat-to-beat contractile performance of the heart is monitored by cardiopulmonary baroreceptors (Kircheim, 1976) and 247 by intracardiac mechanoreceptors (Donald, 1979). Those mechanoreceptors located in the atrial and ventricular walls respond to the rate and force of cardiac contraction, and are linked by vagal afferent fibres to the medullary vasomotor centre where they exert a tonically inhibitory effect (Mancia, 1975). Reductions in the rate and force of ventricular contraction cause disinhibition of the vasomotor centre, increasing cardiac and/or peripheral circulatory outflow (Malliani et al, 1973; Abboud, 1979; Shepherd, 1982) . Propranolol has been shown to reduce ventricular efferent C-fibre activity in cats by reducing heart rate and contractility (Thames et al, 1980; Thoren et al, 1977); this mechanism may explain the increase in cardiac noradrenaline release observed in this study. The absence of increased peripheral sympathetic activity would be explained by a selective increase in cardiac sympathetic drive. This observation is not consistent with the concept of sympathetic activation as a generalised phenomenon, but there is now a considerable body of data which indicates that sympathetic activity may be directed selectively at specific organs in a manner appropriate to physiological requirements. This aspect of peripheral circulatory control has been reviewed in detail by Abboud (1979). The present results support the view that sympathetic efferent activity to various organs may be nonuniform. 12.5.2 Effects of propranolol on coupled pacing 248 It is more difficult to interpret the effects of propranolol on the changes in cardiac and peripheral noradrenaline which occurred during coupled pacing. The previous study (chapter 11) showed that coupled pacing increased both cardiac and circulating noradrenaline release, possibly due to activation of a baroreflex mechanism. Myocardial oxygen consumption and coronary blood flow also increased. These effects were modified when coupled pacing was repeated in the presence of propranolol. The rise in plasma noradrenaline was abolished. Cardiac noradrenaline release also failed to rise (although this is more difficult to interpret because of the shift in baseline values). However, these results would be in keeping with propranolol-induced attenuation of the sympathetic neural response to coupled pacing. A number of animal studies suggest that propranolol can reduce sympathetic outflow at various levels of the reflex arc (Chevalier-Cholat et al, 1978; Tuttle and McCleary, 1978; Korner, 1980; Lewis and Haeusler, 1975). Recently, in man, Ferguson et al (1983) have shown that propranolol attenuates the forearm vasoconstrictor response to baroreflex stimulation. The underlying mechanisms are currently unknown, although a reduction in afferent impulses of cardiac origin may be a common factor. Mueller and Ayres (1980) showed that propranolol lowered plasma noradrenaline levels in patients with acute myocardial infarction and considered a peripheral mechanism, namely blockade of presynaptic beta-2 receptors, as a possible 249 explanation. This is supported by some recent evidence that presynaptic regulation of neurotransmitter release may have effects in vivo (Vincent et al, 1982; Nezu et alf 1985). These possible mechanisms would not be mutually exclusive. Further work with selective beta-1 and beta-2 antagonists is required to assess the importance of presynaptic receptors in man. It may be that the effects of propranolol on the sympathetic nervous system in the resting state differ from its effects during periods of sympathetic activation. At rest, the negative inotropic effect of acute beta blockade appears to provoke a compensatory increase in cardiac sympathetic drive which presumably minimises the reduction in left ventricular performance. However, when sympathetic activity has been stimulated by some other factor, propranolol exerts a dampening effect by mechanisms which have not been fully explained. 12.5.3 Role of catecholamine uptake The ability to estimate cardiac catecholamine uptake and release, seperately but simultaneously in vivo may prove a useful technique in elucidating the sympathetic physiology of the heart. Studies of the neuronal uptake of catecholamines in man have confirmed the physiological importance of this mechanism for terminating the effects of circulating and neuronally-released catecholamines (Goldstein et al, 1988) Human studies confirm that while adrenaline is invariably extracted from coronary blood, there may be either net production or net 250 extraction of noradrenaline (Schwartz et al, 1979; Dominiak et al, 1985; Rose et al, 1983), presumably depending on the rate of release. Whether the rate of catecholamine uptake in vivo remains fixed, or varies, is currently unknown. In the present study, adrenaline uptake increased after propranolol, but not at any other time. The significance of this single observation is not clear. However, in patients with previous left ventricular failure, Rose and colleagues found that adrenaline uptake was reduced or absent; and this was associated with a highr net overflow of noradrenaline. In contrast, a group without left ventricular failure had normal adrenaline uptake (46%) and a small net extraction of noradrenaline (Rose et al, 1983). It is likely that these estimates reflect disturbed sympathetic function in the failing heart since they are consistent with reports of increased sympathetic activity in the failing heart in vivo (Chidsey et al, 1965; Thomas et al, 1978) . Furthermore, studies in vitro on failing myocardial tissue have shown impaired catecholamine uptake (Petch and Nayler, 1979) and reduced catecholamine fluoresence in the storage granules of sympathetic neurones (Vogel et al, 1969). Further studies will be required to assess the role of catecholamine uptake in cardiac failure, and to determine its relationship with changes in noradrenaline release and impaired cardiac function. 251 CHAPTER 13 CONCLUSIONS 252 The relationship between increasing sympathetic activity and the consequent increases in plasma noradrenaline concentrations is linear for a given stimulus, provided that there are no major haemodynamic changes involved. Different sympathetic responses may be associated with different regional patterns of sympathetic activity; therefore, even within an individual, the relationship between sympathetic outflow and plasma noradrenaline may vary with different stimuli. The smallest increase in basal plasma noradrenaline that can be detected with confidence is about 35-40%. However, because regional symapathetic outflow has variable effects on plasma noradrenaline, the significance of modest changes in plasma noradrenaline is uncertain. Peripheral venous plasma noradrenaline levels are a useful index of sympathetic activity within a group in which a single response is being studied. However, the validity of plasma noradrenaline levels as a means of comparing modest differences in sympathetic activity between groups is doubtful. Substantial differences in plasma noradrenaline may reflect true differences in sympathetic activity, but should be considered as a guide and not as a quantitative measure. Assessment of cardiac sympathetic activity can be achieved using invasive vascular techniques. This can provide 253 an insight into noradrenaline uptake and release. The technique of coupled pacing elicited cardiac haemodynamic and metabolic responses consistent with sympathetic stimulation, and this was accompanied by increased cardiac noradrenaline release. The effects of propranolol were complex. At rest, the administration of propranolol increased noradrenaline release in the heart, but appeared to attenuate the sympathetic response to coupled pacing. 254 ACKNOWLEDGEMENTS It is with pleasure that I acknowledge the advice, encouragement and help of Professor Colin Dollery and Dr Morris Brown who introduced me to this fascinating area of physiology and helped to make sense of it. Roger Causon is owed a special debt of gratitude for baling me out of so many tricky spots during the interminable assays, and Dr Michael Murphy as well for giving me so much assistance in the clinical lab. I am grateful to all the members of the departments of cardiology and clinical pharmacology in Creteil for making me so welcome and for going out of their way to understand my very approximate French. In particular Professeur Francois Lhoste who helped arrange the transfer, and Dr Herbert Geschwind who helped my projects to dovetail into the studies taking place in the catheter lab. Dr Jean-Francois Dhainault did most of the catheter placing on my behalf, and Catherine Sabatier virtually all the assays. 255 APPENDIX The catecholamine assay instruction sheet is enclosed. 256 UNIVERSITY of LON D O N Royal Postgraduate Medical School Hammersmith Hospital. Ducane Road, London W12 OHS Tel: 01-743 2030 Ext 3 .Q 5 M -______Department of Clinical Pharmacology Director Professor C T Dollety. BSc, MB, ChB. FRCP PLASMA CATECHOLAMINE RALIOENZYMATIC ASSAY ( 2 4 / 4 / 8 1 ) Arrange small plastic tubes in triplicate - "vertically" in 3-rowed racks and place in an ice-bath. Pipette 3x 50ul plasma samples into each set of 3 tubes. Each assay should include 1 blank per rack and 3-4 internal standards per assay ( a typical non haemolysed plasma + 5ul of lOOng/ml NA and 5 0 n g / m l A ) and 1 pooled plasma triplicate. INCUBATION MIXTURE : - E or 2 r a c k s 3 mg G .S .H . 1 .4 ml Tris/Mg/EG-TA Buffer pH 8.4 4 mg COMT p r e p a r a t io n 10 ul Benzylhydroxylamine, 5mg/ml DIVIDE INTCK 2 JPARTS (A) 0.775 ml 3 “(B) 0.625 ml , 14 Add 300ul 65-85Ci/mmol H-SAM Add lOul 57.6mCi/mmolC-SAM Add 25ul NA lOug/ml Add 25ul A lOug/ml Q Ada 25 ul of (A) to the first 2 rows ( H tubes ) of each rack. Add 25 ul of ( b ) to the third row 04c tubes ) of each rack. INCUBATE EOR 1 .5 HOURS IN 2 5 °C WATER BATH During incubation :- put away plasma samples and G.S.H., prepare T.L.C. tanks, prepare glass tubes for cold carriers, prepare large plastic tubes for NH40K. AT END OP INCUBATION PLACE TUBES IN ICE-BATH To the 14C tubes ONLY add 20ul cold SAM 2mg/ml 1M Borate pH 8 B riefly and gently vortex the 14C tubes. VERY CAREPULLY pipette 40ul from the 14C tubes to each of the 3H tubes, discarding the 14C tubes as you go. Gently vortex the remaining 3H tubes. PREPARE STOPPING MIXTURE P or 3 r a c k s v 15 mg Tetraphenylboron — ...... —' 3 m3. 1M B o r a te pH 8 25 ul cold carriers ( 1 mg/ml ) Add stopping mixture ONLY when ready to proceed to next step UNIVERSITY of LON D O N Royal Postgraduate Medical Schoo Hammersmith Hospital, Ducane Road, London W12 OHS Tel: 01-743 2030 Ext 2 o S t j -______Department of Clinical Pharmacology Director Professor C T Dollery, BSc. MB. ChB. FRCP 3 Add 50ul of stopping mixture to the H tubes. Dispense 2 ml diethyl ether into each tube. Vortex for 2 min, at no.24- Centrifuge at 2,000 rpm for 2 min. Freeze lower aqueous layer in solid C02/ acetone mixture. Tip upper diethyl ether layer into glass tubes containing 25ul cold carriers ( 1 mg/ml in 0. 1 M HC1 ). Vortex for 3 min. at no. 30* Centrifuge for 2 min, at 2,000 rpm. Freeze lower aqueous layer in solid CO 2 / acetone mixture. Discard upper diethyl ether layer, leaving pellet in tube. Blow off residual "ether” in Bucliler Vortex Evaporator, 90 seconds at room temperature. Spot onto Y/hatman LKD 5F 19-channel TIC plates, Leave for 10 min, then blow dry in COLD air for 15 min. Run the plates in chloroform,methanol,70^ethylamine :- 32 : 6 : 4• ( ~ 1 hr ). Remove plates, dry briefly in fume-cupboard, visualise under Ultra-Violet light and locate spots using a scalpel. solvent front 3-MT ( DA ) MN ( A ) NMN ( NA ) Moisten each spot with a drop of distilled water, then scrape into large plastic tubes containing 1ml 0.05M NH40H. Vortex for 10 min. at no. 20. Add 7 5 u l FRESHLY PREPARED 4i° sodium-m-periodate ( store on ice ) Briefly vortex and start clock, add 10 ml of toluene followed by 25ul glacial acetic acid at 10 min. Extract for 2 min. by inversion. Freeze lower aqueous layer in solid CO 2 / acetone mixture, tip upper toluene layer into scintillation vials containing O .4 ml Permafluor. Count using UR4d for NA and UR47 for A. - — BIBLIOGRAPHY Abboud F. Integration of reflex responses in the control of blood pressure and vascular resistance. Am J Cardiol 1979; 44: 903-911. Adler-Graschinsky E, Langer SZ. Possible role of a beta adrenoceptor in the regulation of noradrenaline release by nerve stimulation through a positive feedback mechanism. Br J Pharmacol 1975; 53: 43-50. Ahmet SS, Levinson GE, Schwarz CJ, Ettinger PD. Systolic time intervals as measures of the contractile state of the left ventricular in man. Circ 1972; 46: 559-571. Ahiquist RP. A study of the adrenotropic receptors. Am J Physio 1948; 153: 506. Axelrod J. Purification and properties of phenylethanolamine-N-methyltransferase. J Biol Chem 1962; 237: 1657-1660. Axelrod J. Dopamine beta hydroxylase: regulation of its synthesis and release from nerve terminals. Pharmacol Rev 1972; 23: 233-244. Barger G, Dale HH. Chemical structure and sympathomimetic action of amines. J Physiol 1910; 41, 19-59. Beck W, Chester E, Schrie V. Postextrasystolic ventricular pressure responses. Circulation 1971; 44: 523-527. Bello CT, Sevy RW, Harakal C. Varying haemodynamic patterns in essential hypertension. Am J Med Sci 1965; 250: 24-35. Bello CT, Sevy RW, Harakal C, Hillyer PN. Relationship between clinical severity of disease and haemodynamic patterns in essential hypertension. Am J Med Sci 1967; 253: 194-208. Bevan JA, Su C. Variation of intra- and perisynaptic adrenergic transmitter concentrations with width of synaptic cleft in vascular tissue. J Pharmacol Exp Ther 1974; 190: 30-38. Bevan JA. Some functional consequences of variation in adrenergic synaptic cleft width and in nerve density and distribution. Fed Proc 1977; 36: 2439. 257 Bevan JA. Some bases in differences in vascular response to sympathetic activity. Circ Res 1979; 45: 161-171. Blashko H. Catecholamine biosynthesis. Brit Med Bull 1973; 29: 105-109. Blaustein MP. The role of calcium in catecholamine release from adrenergic nerve terminals. In: The release of catecholamines fron adrenergic neurons. Paton DM, Ed. Pergamon Press, Oxford. 1979, ps 39-58. Boerth R, Hammermeister KE, Warbasse JR. Comparative influence of norepinephrine and heart rate on myocardial oxygen consumption and inotropic state in dogs. Am Hear J 1978; 95: 355-361. Bonelli J, Hartnaegl H, Bruecke TH, Magometschnigg D, Lochs H, Kaik G. Effects of calculation stress on haemodynamic and plasma catecholamines before and after beta-blockade with propranolol and mepindolol sulphate. Eur J Clin Pharmacol 1979;15: 1-8. Braunwald NS, Gay WA, Morrow AG, Braunwald E. Sustained paired electrical stimuli. Slowing of the ventricular rate and augmentation of the contractile force. Am J Cardiol 1964; 14: 385-395. Bravo EL, Tarazi RC. Plasma catecholamines in clinical investigation: a useful index or a meaningless number? J Lab Clin Med 1982; 100:155-160. Brect HM, Banthien F, Ernst W, Schoeepe W. Increased plasma noradrenaline concentrations in essential hypertension and their decrease after treatment with a beta-receptor blocking agent (prindolol). Clin Sci 1976; 51: 485s. Bristow MR, Ginsburg R, Minobe W. et al. Decreased catecholamine sensitivity and beta adrenergic receptor density in failing human hearts. N Engl J Med 1982; 307: 205-211. Brown GL, Davies BN, Ferry CB. Effect of neuronal rest on the output of sympathetic transmitter from the spleen. J Physiol 1961; 159: 365-378. Brown GL, Gillespie JS. The output of sympathetic transmitter from the spleen of the cat. J Physiol 1957; 138: 81-88. Brown MJ, Dollery CT. Adrenaline and hypertension. Clin Exp Hyper 1984; 8 6 : 539-549. Brown MJ, Jenner DA. A novel double-isotope technique for ensythetic assay of plasma catecholamines, permitting high precision, sensitivity and sample capacity. Clin Sci 1981; 61: 591-598. 258 Brown MJ, Jenner DA, Alison DA, Dollery CT. Variations in individual organ release of noradrenaline measured by an improved radioenzymatic technique; limitations of peripheral venous measurements in the assessment of sympathetic activity. Clin Sci 1981; 61; 535-590. Brown MJ, Lhoste FJM, Zamboulis C, Ind PW, Jenner DA, Dollery CT. Estimation of sympathetic activity in e ss e n tia l hypertension. Clin Pharmacol Ther 1982; 31: 16-22. Burgen ASV, Iversen LL. Inhibition of noradrenaline uptake by sympathomimetic animines in the rat isolated heart. Brit J Pharmacol 1965; 25: 34-49. Burke D, Sundlof G, Wallin BG. Postural effects on muscle nerve sympathetic activity in man. J Physiol 1977; 272: 399. Cannon WB. Organisation for physiological homeostasis. Physiol Rev 1929; 9, 399-431. Carruthers M, Taggart P, Somerville W. Plasma-catecholamine estimations. Lancet 1970; 2: 421. Chang PC, van der Koot JA, Vermeij P, van Brummelen P. Norepinephrine removal and release in the forearm of healthy subjects. Hypertension 1986; 8 : 801-809. Chevalier-Cholat AM, Friggi A, Torresani J. Action of beta-adrenergic blocking agents on haemodynamic variables and on baroreceptor and postganglionic activities in rabbits. Biomedicine 1978, 28: 67-72. Chidsey CA, Braunwald E, Morrow AG. Catecholamine excretion and cardiac stores of norepinephrine in congestive heart failure. Am J Med 1965; 39: 442-451. Christensen NJ. Plasmas catecholamines in long-term diabetics with and without neuropathy and in hypophysectomized subjects. J Clin Invest 1972; 51: 779. Christensen NJ, Videbaek J. Plasma catecholamines and carbohydrate metabolism in patients with acute myocardial infarction. J Clin Invest 1974; 54: 278-286. Cohn PF. Evaluation of inotropic contractile reserve in ischaemic heart disease using postextrasystolic potentiation. Circulation 1980; 61: 1071-1075. Cousineau D, Ferguson RJ, de Champlain J, Gauthier P, Cote P, Bourassa M. Catecholamines in coronary sinus during exercise in man before and after training. J Appl Physiol 1977; 43: 801-806. 259 Cousineau D, Rose CP, Goresky CA. Labelled catecholamine uptake in the dog heart. Circ Red 1980; 47: 329-338. Cryer PE. Physiology and pathophysiology of the human sympathoadrenal neuroendocrine system. N Engl J Med 1980; 303: 436-444. Cryer PE, Haymond MW, Santiago SV, Such SD. Norepinephrine and Epinephrine release and adrenergic mediation of smoking associated haemodynamic and metabolic events. N Engl J Med 1976; 295: 573-577. Cryer PE, Rizza RA, Haymond MW, Gerich JE. Epinephrine and norepinephrine are cleared through beta-adrenergic but not alpha-adrenergic mechanisms in man. Metabolism 1980; 29: 1114-1118. Cryer PE, Hammond MW, Santiago JV, Shah SD. Norepinephrine and epinephrine release and adrenergic mediation of sm oking-associated haemodynamic and m etabolic even ts. N Engl J Med 1976; 295: 573-577. Dale HH. Tha actions of certain esters and ethers of choline, and their relation to muscarine. J Pharmac exp Ther 1914; 2:147-190. Dale HH. Nomenclature of fibres in the autonomic nervous system and their effects. J Physiol 1933; 80: 10-11. DaPrada M, Zurcher G. Simoultaneous radioenzymatic determination of plasma and tissue adrenaline, noradrenaline and dopamine within the femtomole range. Life Sci 1976; 19: 1161-1174. Dargie HJ, Dollery CT, Lewis PJ. Prevention of DOCA-salt hypertension by central 6 -hydroxydopamine: role of saline intake. Brit J Pharmacol 1975; 53: 455P. Davignon A, Rey C, Payot M et al. In: Juvenile Hypertension ppl89-193, Ed: New MI and Levine LS. Raven Press, New York. Day MD. Autonomic pharmacology - experimental and clinical aspects. 1979. Churchill Livingstone, Edinburgh. De Champlain J, Krakoff LR, Axelrod J. The metabolism of norepinephrine in experimental hypertension in rats. Circ Res 1967; 20: 136-148. De Champlain J, Krakoff LR, Axelrod J. Interrelation of sodium intake, hypertension and norepinephrine storage in the rat. Circ Res 1969; 24 (suppl I): 75-90. 260 De Champlain J, van Ameringen MR. Regulation of blood pressure by sympathetic fibers and adrenal medulla in normotensive and hypertensive rats. Circ Res 1972; 31: 617-626. De Champlain J, Farley L, Cousineau D, van Ameingen MR. Circulating catecholamine levels in human and experimental hypertension. Circ Res 1976; 39: 109-116. De Champlain J, Cousineau D, van Ameringen MR, Marc-Aurele J, Yamagachi N. The role of the sympathetic system in experimental and human hypertension. Postgrad Med J 1977; 53 (suppl 3): 15-30. Delius W, Hagbarth KE, Hongell A, Wallin BG. Manouevres affecting sympathetic outflow in human muscle nerves. Acta Physiol Scand 1972; 84: 82-89. DeQuattro V, Campese V, Lurvey A, Yen G, Kypridakis G. Low response of serum dopamine beta hydroxylase to stimuli in primary hypertension. Biochem Med 1976; 15: 1-9. DeQuattro V, Campese V, Miura Y, Meijer D. Increased plasma catecholamines in high reini hypertension. Am J Cardiol; 38: 801-812. DeQuattro V, Chan S. Raised plasma catecholamines in some patients with primary hypertension. Lancet 1972; 1: 806-809. Dixon WE. On the mode of action of drugs. Medical Magazine (London) 1907; 16:454-457. Dominiak P, Delius W, Grobecker H. Changes in plasma catecholamine contractions in patients with coronary heart disease during pacing and physical exercise. Clin Cardiol 1985; 8 : 77-81. Donald DE, Shepherd JT. Cardiac receptors: normal and disturbed function. Am J Cardiol 1979; 44: 873-878. Dunstan HP, Tarazi RC. In: Juvenile Hypertension, ppl81-188, Ed: New MI & Levine LS. Raven Press, New York. Elliott TR. The action of adrenalin. J Physiol 1904; 401-467. Elliott HL, Reid JL. Evidence for postjunctional alpha adrenoceotors in peripheral vascular regulation in man. Clin Sci 1983; 65: 237-241. Engelman K, Portnoy B, Sjoerdsma A. Plasma catecholamine concentrations in patients with hypertension. Circ Res 1970; 26 and 7 (suppl 1): 1-141. 261 Epstein SE, Stampfer M, Beiser GD, Goldstein RE, Braunwald E. Effects of a reduction in environmental temperature on the circulatory response to exercise in man. New Engl J Med 1969; 280: 7-11. Esler MD, Hasking GJ, Willett IR, Leonard PW, Jennings GL. Noradrenline release and sympathetic nervous system activity. J Hypertension 1985; 3: 117-129 Esler M, Jackman G, Bobik A, Kelleher D, Jennings G, Leonard P, Skews H, Korner P. Determination of norepinephrine apparent release rate and clearance in humans. Life Sci 1979; 25: 1461-1470. Esler M, Jackman G, Leonard P, Skews H, Bobik A, Jennings G. Effect of propranolol on noradrenaline kinetics in patients with essential hypertension. Br J Clin Pharmac 1981; 12: 375-380. Esler M, Nestel PJ. Renin and sympathetic nervous system responsiveness to adrenergic stimuli in Essential hypertension. Am J Cardiol 1973; 32:643-649. Esler M, Skews H, Leonard P, Jackman G, Bobik A, Korner P. Age-dependence of noradrenaline kinetics in normal subjects. Clin Sci 1981; 60: 217-219. Esler M. Assessment of sympathetic nervous function in humans from noradrenaline plasma kinetics. Clin Sci 1982; 62: 247-254. Esler M, Zweifler A, Randall 0, Julius S, Bennett J, Rydelek P. Suppression of sympathetic nervous function in low renin essential hypertension. Lancet 1976; 2: 115-118. Esler M, Zweifler A, Randall O, Julius S, DeQuattro V. Agreement among three different indices of sympathetic nervous system activity in essential hypertension. Mayo Clin Proc 1977; 52: 379-386. Falck B, Hillarp NA, Thieme G, Torp A. Fluorescence of catecholamines and related compounds condensed with formaldehyde. J Histochem Cytochem 1962; 10: 348-354. Ferguson DW, Thames MD, Mark AL. E ffects of propranolol on reflex vascular responses to orthostatic stress in humans. Circulation 1983; 67: 802-807. Fitzgerald GA, Hossman V, Dollery CT. Norepinephrine release in essential hypertension. Clin Pharmacol Ther 1981; 30: 164-171. Fitzgerald GA, Hossmann V, Hamilton CA, Reid JL, Davies DS, Dollery CT. Interindividual variation in kinetics of infused norepinephrine. Clin Pharmacol Ther 1979; 26: 669-675. 262 Fillenz M. Ultrastructural studies of the mechanism of release. In: The release of catecholamines from adrenergic neurons, Ed. Paton DM. Pergamon Press, Oxford. 1979 ps 17-37 Flammer J, Weidman P, Gluck Z, Ziegler WH, Reubi FC. Cardiovascular and endocrine profile of adrenergic neurone blockade in normal and hypertensive man. Am J Med 1979; 6 6 : 34. Folkow B, Bona GF, Hjemdahl P, Toren PH, Wallin G. Measurements of plasma norepinephrine concentratin sin primary hypertension. A word of caution of their applicability for assessing neurogenic contributions. Hypertension 1983; 5: 399-403. Forman MB, Robertson D, Goldberg M, Bostick D, Uderman H, Perry TM and Robertson RM. Effect of tyramine on mycocardial catecholamine release in coronary heart disease. Am J Cardiol 1984; 53: 476-480. Freis ED. Haemodynamics of Hypertension. Physiol Rev 1960; 40: 27-54. Frohlich ED, Tarazi RC, Dunstan HP. Re-examination of the haemodynamics of hypertension. Am J Med Sci 1969; 257: 9-23. Galbo H, Holst JJ, Christensen NJ, Hilsted J. Glucagon and plasma catecholamines during beta-receptor blockade in exercising man. J Appl Physiol 1976; 40: 855-863. Ganz W, Tamura K, Marcus H, Donoso R, Yoshida S, Swan HJC. Measurement of coronary sinus blood flow by continuous thermodilution in man. Circulation 1971; 44: 181-185. Garber AJ, Cryer PE, Santiago JV, Haymond MW, Pagliara AS, Kipuis DM. The role of adrenergic mechanisms in the substrate and harmonal response to insulin induced hypoglycaemia in man. J Clin Invest 1976; 58: 7. Gauthier P, Nadeau RA, de Champlain J. Cardiovascular reactivity in the dog after chemical sympathectomy with 6 -hydroxydopamine. Can J Physiol Pharmacol 1978; 52: 590-601. Geffen LB, Livett BG. Synaptic vesicles in sympathetic neurons. Physiol Rev 1971; 57: 98-159. Geffen LB, Rush RA, Louis WJ, Doyle AE. Plasma dopamine beta-hydroxylase and noradrenaline amounts in essential hypertension. Clin Sci 1973; 44: 617. 263 Geshwind HJ, Lhoste F, Scriven AJI, Dhainaut JF, Sabatier C, Laurent D. Sympathetic nervous system activation in postextrasystolic potentiation: Role of catecholamine release in enhancement of ventricular function. J Am Coll Cardiol 1984; 4: 216-225. Ghione S, Palombo C, Pellegrini M, Formmie E, Pilo A, Donato L. The kinetics of plasma noradrenaline in normal and hypertensive subjects. Clin Sci Mol Med 1978; 55: 895-925. Gibson DG. Use of the systolic time internals in clinical pharmacology. Br J Clin Pharmacol 1978; 6 : 97-102. Gillespie JS. Uptake of noradrenaline by smooth muscle. Br Med Bull 1973; 29: 136-141. Glowinski J, Kopin IJ, Axelrod J. Metabolism of 3H norepinephrine in the rat brain. J Neurochem 1965; 12: 25. Goldstein DS. Plasma norepinephrine in essential hypertension. A study of studies. Hypertension 1981; 3: 48. Goldstein DS. Plasma catecholamines and essential hypertension. An analytical review. Hypertension 1983; 5:86-99. Goldstein DS, Brush JE, Eisnhofer G, Stull R, Esler MD. In vivo measurement of neuronal uptake of norepinephrine in the human heart. Circulation 1988; 78:41-48. Goldstein DS, Dionne R, Sweet J, Gracely R, Brewer HB Jr, Gregg R, Keiser HR. Circulatory plasma CA, cortisol, lipid and psychological responses to a real-life stress (wisdom tooth extractions): effects of diazepan sedation and of an inclusion of epinephrine with the local anaesthetic. Psychosom Med 1982; 44: 259. Goldstein DS, McCarty R, Policisky RJ, Kopin J. Relationship between plasma norepinephrine and sympathetic neural activity. Hypertension 1983; 5: 552-559. Goldstein DS, Zimlichman R, Stall R, Keiser HR, Kopin IJ. Estimation of intrasynaptic norepinephrine in humans. Hypertension 1986; 8 : 471-475. Grimm M, Weidmann P, Keusch G, Meier A, Gluck. Norepinephrine clearance and pressure effect in normal and hypertensive man. Klin Wochenschr 1980; 58: 1175-1181. Gutteberg T, Borud 0, Stromme JH. Dopamine beta hydroxylase activity in serum following acute myocardial infarction; an evaluation of this parameter for routine use as an index of sympathetic activity. Clin Chim Acta 1976; 69: 61-66. 264 Haefely W, Hurlimenn A, Thoenen H. Relationship between the rate of stimulation and the quantity of noradrenaline liberated from sympathetic nerve endings in the isolated perfused spleen of the cat. J Physiol (Lond) 1965; 181: 48-58. Haneda T, Miura Y, Arai T, Nakajima T, Miura T, et al. Norepinephrine levels in the coronary sinus in patients with cardiovascular diseases at rest and during isometric handgrip exercise. Am Heart J 1980; 100: 465-472. Hansen L, Zweifley AJ, Julius S, Hunyor S. Haemodynamic effects of acute and prolonged beta-adrenergic blockade in essential hypertension. Acta Med Scand 1974; 196: 27-34. Hattenhauer M, Neill WA. The effect of cold air inhalation on angina pectoris and myocardial oxygen supply. Circulation 1975; 51: 1053-1058. Hausen LF, Hesse B, Christensen NJ. Enhanced sympathetic nerve activity after intravenous propranolol in ischaemic heart disease: plasma noradrenaline, splanchnic blood flow and mixed venous oxygen saturation at rest and during exercise. Eur J Clin Invest 1978; 8 : 31-36. Henquet JW, van Baak M, Schols M, Rahn KH. Studies on the autonomic nervous system in borderline hypertension. Eur J Clin Pharmacol 1982; 22: 285-288. Henry DP, Starman BJ, Johnson DG, Williams RH. A sensitive radioenzymatic assay for norepinephrine in tissues and plasma. Life Sci 1975; 16: 375-384. Hertling G, Axelrod J, Whitby LG. Effect of drugs on the uptake and metabolism of 3H-NE. J Pharmacol Exp Ther 1961; 134: 146-153. Hillarp NA. The construction and functional organisation of the automatic innervation apparatus. Acta Physiol Scand 1959; 46: Suppl. 157: 1-39. Hirshfield JW, Epstein SE, Roberts AJ. Indices predicting long term survival after valve replacement in patients with aortic regurgitation and patients with aortic stenosis. Circulation 1974; 50: 1190-1199. Hjemdahl P, Eklund B, Kaijser L. Catecholamine handling by the human forearm at rest and during isometric exercise and lower body negative pressure. Br J Pharmacol 1982; 77: 324p. Hjemdahl P, Eliasson K. Sympatho-adrenal and cardiovascular response to mental stress and orthostatic provocation in latent hypertension. Clin Sci 1979; 57 (Suppl): 189s. 265 Hoeldtke RD, Cilmi KM, Reichard GA, Boden G, Owen Oe. Assessment of norepinephrine secretion and production. J Lab Clin Med 1983, 101: 772-782. Hoertnagl H, Winkler H, Lochs H. Membrane proteins of chromaffin granules, dopamine beta hydroxylase a major constituent. Biochem J 1972; 129: 187-195. Horwitz D, Alexander RW, Lovenberg W. Human serum dopamine beta hydroxylase. Relationship to hypertension and sympathetic a c tiv ity . Circ Res 1973; 32: 594-599. Hoffman BF, Bartlestone HJ, Scherlag RJ, Cranefield PF. Effects of postextrasystolic potentiation on normal and failing hearts. Bull NY Acad Med 1965; 41: 498-503. Huet Y, Geschwind H, Laine JF, Comoy E, Laurent D. The energetic cost of postextrasystolic potentiation. Circulation 1981: 4 (suppl IV); IV325. Iversen LL. The uptake and storage of noradrenaline in sympathetic neurones. Oxford: Cambridge University Press. 1967: 114-121. Iversen LL. Catecholamine uptake processes. Brit Med Bull 1973; 29: 130-135. Jie K, van Brummelen P, Vermey P, Timmermans PBMWM, van Zwieten P. Differences between exogenous and endogenous noradrenaline in the effects on vascular post-synaptic alpha-1 and alpha-2 adrenoceptors in man. J Hypertens 1985; 3 (suppl 3): sl45-sl47. Johnson DG, Hayward JS, Jacobs TP, Collis ML, Eckerson JD, Williams RH. Plasma norepinephrine and responses of man in cold water. J Appl Physiol 1977; 43: 216-220. Jones DH, Hamilton CA, Reid JL. Choice of control groups in the appraisal of sympathetic nervous activity in essential hypertension. Clin Sci 1979; 57: 339. Julius S, Conway J. Hemodynamic studies in patients with borderline blood pressure elevation. Circ 1968; 38: 282-288. Julius S, Pascual AV, Reilly K et al. Relationship between cardiac output and peripheral resistance in borderline hypertension. Circ 1971; 43: 382-390. Julius S, Pascual AV, Sannerstedt R and Mitchell C. Relationship between cardiac output and peripheral resistance in borderline hypertension. Circ 1971; 43: 382-390. 266 Julius S, Esler M. Automatic nervous cardiovascular regulation in borderline hypertension. Am J Cardiol 1975; 36: 685-695. Karlsberg RP, Penkoske PA, Cryer PE, Corr PB, Roberts R. Rapid activation of the sympathetic nervous system following coronary artery occasion : relationship to infarct size, site and hemodynamic impact. Cardiovasc Res 1979; 13: 523-531. Kircheim HR. Systemic arterial barroreceptor reflexes. Physiol Rev 1976; 56: 100-176. Kopin IJ. Biochemical assessment of peripheral adrenergic activity, In: The release of from Adrenergic Neurones, pp 353-372, Ed: Patol DM, Pergannon Press, Oxford. Kopin IJ. Plasma catecholamines in human and experimental hypertension. In Nervous System and Hypertension, edited by Meyer P, Schmitt H, New York. Wiley 1979; 267-276. Korner PI, Dorward PK, Blomery PA, Frean GJ. Central nervous beta-adrenoceptors and their role in the cardiovascular action of propranolol in rabbits. Circ Res 1980; 46 (suppl 1): 1-26. Kozlowski S, Brzezinski A, Nazar K, Kowalski W, Franczyk M Plasma catecholamines during sustained isometric exercise. Clin Sci Mol Med 1973; 45: 723-727. Krakoff LR, de Champlain J, Axelrod J. Abnormal storage of norepinephrine in experimental hypertension in the rat. Circ Res 1967; 21: 583-590. Kuramoto K, Murata K, Yazaki Y et al. Hemodynamics in the juvenile hypertension with special reference to the response of propranolol. Jap Circ 1968; 32: 981-987. Kvetnansky R, Weise VK, Thoa NB, Kopin IJ. The effects of chronic guanethidine therapy and adrenal medullectomy on plasma catecholamines and cortisterone in forcibly immobilised rats. J Pharmacol Exp Ther 1979; 209: 287. Lake CR, Ziegler MG, Kopin IJ. Use of plasma norepinephrine for evaluation of sympathetic neuronal function in man. Life Sci 1976; 18: 1315. Lake CR, Ziegler MG, Coleman MD, Kopin IJ. Age-adjusted plasma norepinephrine levels are similar in normotensive and hypertensive subjects. N Engl J Med 1977; 296: 208-214. Lands AM, Luduena FP, Buzzo HJ. D iffer en tia tio n of receptor systems responsive to isoproterenol. Life Sciences 1967; 6 : 2241-2252. 267 Langer SZ. The metabolism of 3H-NA released by electrical stimulation from the isolated nictitating membrane of the cat and from the vas deferens of the rat. J Physiol (Lond) 1970; 208: 515-546. Langer SZ. Presynaptic receptors and their role in the regulation of transmitter release. Br J Pharmacol 1977; 60: 481-497. Langer SZ. Presynaptic adrenoceptors and regulation of release. In: The release of catecholamines from adrenergic neurons Ed: D M Paton 1979; 59. Langer SZ. Presynaptic regulation of the release of catecholamines. Pharmacol Rev 1980, 32: 337-362. Langer SZ, Carero I, Massingham R. Recent developments in noradrenergic neurotransmission and its relevance to the mechanism of action of certain antihypertensive agents. Hypertension 1980; 2: 372-382. Langer SZ, Massingham R, Shepperson N. Presence of postsynaptic alpha 2 adrenoceptors of predominantly extra synaptic location in the vascular smooth muscle of the dog hind limb. Clin Sci 1980; 59: 2255-2285. Langley JN. Observations on the physiological actions of the supra-renal bodies. J Physiol 1901; 27: 237-256. Langley JN. Croonian Lecture. On nerve endings and on special excitable substances in cells. Proc R Soc B 1906; 78; 170-194. LeBlanc J, Cote J, Jobin M, Labrie A. Plasma catecholamines and cardiovascular responses to cold and mental activity. J Appl Physiol 1979; 47: 1207-1211. Lee RMKW, T riggle CR, Cheung DWT, Coughlin MD. Structural and functional consequence of neonatal sympathectomy on the blood vesels of spontaneously hypertensive rat. Hypertension 1987; 10: 328-338. Leitz FH, Stefano FSE. The effect of tyramine, amphetamine and metaraminol on the metabolic disposition of 3H-NE released from the adrenergic neuron. J Pharmacol Exp Ther 1971; 178: 464-473. Lendrum B, Feinberg H, Boyd E, Katz LN. Rhythm effects on contractility of the beating isovolumic left ventricle. Am J Physiol 1960; 199: 1115-1121. Levy NM, Blattberg B. The effect of the pattern of cardiac sympathetic activity on myocardial contractile force and norepinephrine overflow in the dog heart. Circ Res 1976; 39: 341-348. 268 Levy MN, Blattberg B. Effect of vagal stimulation on the overflow of norepinephrine into the coronary sinus during cardiac sympathetic nerve stimulation in the dog. Circ Res 1976; 38: 81-85. Lewis PJ, Haeusler J. Reduction in sympathetic nervous activity as a mechanism for the hypotensive effect of propranolol. Nature 1975; 256: 440. Lewis RP, Rittgers SE, Forester WF, Boudoulas H. A critical review of the systolic time intervals. Circulation 1977; 56: 146-183. Lin MS, McNay SL, Shepherd AMM, Kent Keeton T. Effects of hydralazine and sodium nitroprusside on plasma catecholamines and heart rate. Clin Pharmacol Ther 1983; 34: 474-480. Lindemann JP, Bailey JC, Watanabe AM. Potential biochemical mechanisms for regulation of the slow inward current; theoretical basis for drug action. Am Heart J 1982; 103: 746-756. Loewi O. Ubertragbarkeit der Herznervenwirkung. Pfluegers Arch ges Physiol 1921; 189: 238-242. Louis WS, Doyle AE, Anavekar S. Plasma norepinephrine levels in essential hypertension. N Engl J Med 1973; 288: 599-604. Louis WS, Doyle AE, Anavekar S, Johnson Cl, Geffen B. Plasma catecholamine, dopamine beta hydroxylase, and renin levels in essentialhypertension. Circ Res 1974; 34(suppl I): 57-63. Lund-Johansen P. Hemodynamics in early essential hypertension. Acta Med Scand 1967; Suppl. 482: 1-101. Lund-Johansen P. Hemodynamics in essential hypertension. Clin Sci 1980; 59: 3535-3545. Majewski H, Rand MS, Tung LH. Activation of prejunctional beta-adrenoceptors in rat atria by adrenaline applied exogenously or released as a co-transmitter. Br J Pharmacol 1981; 73: 669-679. Maling TJB, Ferrarra A, Mucklow JC, Reid JL, Hamilton CA, Dollery CT. Blood pressure and plasma noradrenaline during single high-dose beta-adrenoceptor blockade. Eur J Clin Pharmacol 1979; 15: 375-379. Malliani A, Parks M, Tuckett RP, Brown AM. Reflex increases in the heart rate elicited by stimulation of afferent cardiac sympathetic nerve fibres in the cat. Circ Res 1973; 32: 9-14. 269 Mancia G, Donald DE. Demonstration that the atria, ventricles and lungs each are responsible for a tonic inhibition of the vasomotor centre in the dog. Circ Res 1975; 36: 310-318. Mancia G, Leonetti G, Picotti GB, Ferrari A, Galva MD, Gregorini L, Parati G, Pomidossi G, Ravazzani C, Sala C, Zanchetti A. Plasma catecholamines and blood pressure responses to the carotid baroreceptor reflex in essential hypertension. Clin Sci 1979; 57 Suppl: 165s. Mathe AA, Tong EY, Tisher PW. Norepinephrine release fron the lung by sympathetic nerve stimulation. Inhibition by vagus and methacholine. Life Sci 1977; 20: 1425-1430. Messerli FH, de Carvalho JGR, Christie B, Frohlich ED. Systemic and regional hemodynamics in low, normal and high cardiac output borderline hypertension. Circ 1978; 58: 441-448. Miura Y, DeQuattro V. Biochemical evaluation of symapthetic nerve tone in essential hypertension. Jpn Circ J 1975; 39: 583-591. Mueller HS, Ayres SM. Propranolol decreases sympathetic nervous activity reflected by plasma catecholamines during evolution of myocardial infarction in man. J Clin Invest 1980; 65: 338-346. Murphy MB, Brown MJ, Causon RC, Scriven AJI, Dollery CT. The effects of pentolinium on sympathetic activity in hypertensives and normotensive controls. Int J Clin Pharm Res 1983; 3: 227-231. Murphy MB, Scriven AJI, Brown MJ, Causon RC, Dollery CT. The effects of nifedipine and hydralazine-induced hypotension on sympathetic activity. Eur J Clin Pharm 1983; 23, 479-482. Muscholl E. Presynaptic muscarine receptors and regulation of release. In: The release of catecholamines fron adrenergic neurons. Paton DM, Ed. Pergamon Press, Oxford. 1979, ps 87-110. Musgrove DF, Bachman AW, Gordon RD. Adrenaline infusion increases plasma noradrenaline in man. J Hyper 1984; 2 (Suppl. 3): 135-137. Nagoka A, Lovenberg W. Plasma norepinephrine and dopamine beta hydroxylase in genetic hypertensive rats. Life Sci 1976; 19: 29-35. Nandi PS, Spodick DH. Recovery from exercise at ranging work loads. Time course of responses of heart rate and systolic intervals. Brit Heart J 1977; 39: 958-966. 270 Nesto RW, Cohn LH, Collins JJ, Wynne J, Holman L, Cohn PF. Inotropic contractile reserve: a useful predictor of increased 5 year survival and improved postoperative left venticular function in patients with coronary artery disease and reduced ejection fraction. Am J Cardiol 1982; 50: 39-44. Nezu M, Miura Y, Adachi M, Adachi M et al. The effects of epinephrine on norepinephrine release in essential hypertension. Hypertension 1985; 7: 187-195. Nielson SL, Christensen NJ, Olssen N, Lassen NA. Raynaud's Phenomenon: peripheral catecholamine concentrations and the effect of sympathectomy. Acta Clin Scand 1980; 502: 57-69. Oliver JA, Pinto J, Sciacco RR, Cannon PJ. Basal norepinephrine overflow into the renal vein: effect of renal nerve stimulation. Am J Physiol 1980; 239: F371. Ozaki M. Metabolism of monamines in spontaneously hypertensive rat. Jap J Pharmacol 1966; 16: 257-237. Palmero HA, Caeiro TE. Statistical analysis of the different factors which could affect postextrasystolic potentiation in the human heart. Am Heart J 1976; 92: 427-434. Peart WS. The nature of splenic sympathin. J Physiol (Lond) 1949; 108: 491-501. Pedersen EB, Christensen NJ. Catecholamines in plasma and urine of patients with essential hypertension determined by double-isotope derivative techniques. Acta Med Scand 1975; 198: 373-384. Petch MC, Nayler WG. Uptake of catecholamines in human cardiac muscle in vitro. Br Heart J 1979; 41: 336-342. Picotti GB, Carruba MO, Ravazzani C, Cesura AM, Galva MD, Da Prada M. Plasma catecholamines in rats exposed to cold. Eur J Pharmacol 1981; 69: 321-329. Pidgeon J, Miller GAH, Noble MIM, Papadomannis D, Seed WA. The Relationship between the strength of the human heart rate beat and the interval between beats. Circulation 1982; 65: 1404-1410. Planz G. Adrenaline and noradrenaline concentrations in the blood of suprarenal and renal vein in man with normal blood pressure and essential hypertension. Klin Wochfeschr 1978; 56: 1109. Polinsky RJ, Kopin IJ, Ebert MH, Weise V. Pharmacologic distinction of different orthostatic hypotension syndromes. Neurology 1981; 31: 131-147. 271 Popio KA, Gorlin R, Bechtel D, Levine JA. Postextrasystolic potentiation as a predictor of potential myocardial viability; preoperative analyses compared wit studies after coronary bypass surgery. Am J Cardiol 39; 944-953. Rahn KH, Gierlichs G, Planz M, Planz R, Schols M, Stephany MW. Studies on the effects of propranolol on plasma catecholamine levels in patients with essential hypertension. Eur J Clin Invest 1978; 8 : 143-148. Rapoport RM, Takimoto GS, Cho AK. Compartmental analysis of tyramine-induced norepinephrine depletion. Pharmacol 1981; 22: 235-242. Reid JL. Acute and chronic beta receptor blockade with propranolol and the cardiovascular responses to intravenous sodium nitroprusside in the hypertensive rabbit. J Cardiovasc Pharmacol 1979; 1: 403-414. Reid JL, Kopin W. The e ffe c ts of gan glion ic blockade, reserphine and v in b la stin e on plasma catecholam ines and dopamine-beta-hydroxylase in the rat. J Pharmacol Exp Ther 1975; 193: 748. Reid JL, Wing LMH, Mathias CJ, Frankel HZ, N e ill E. The central hypotensive action of clonidine in man: Studies in subjects with traumatic cervical cord section. Eur J Clin Invest 1976; 6 : 3-14. Reid JL, Dean CR, Hamilton C, Watson R, Jones DH, Littler W. Relationships between blood pressure, heart rate and plasma noradrenaline. In: Circulating catecholamines and blood pressure. Birkenhaeger WH, Ed. Bunge, Utrecht. 1978 ps 11-16. Robertson D, F rolich JC, Carr RK, Watson JT, H o llifie ld JW, Shand DG, Oates JA. Effects of caffeine on plasma renin activity, catecholamines and blood pressure. N Engl J Med 1978; 298: 181-186. Robertson D, Johnson GA, Robertson RM, Nies AS, Shand DG, Oates JA. Comparative assessment of stimuli that release neuronal and adrenomedullary catecholamines in man. Circulation 1979; 59: 637-643. Rose CP, Burgess JH, Cousineau D. Reduced aortocoronary sinus extraction of epinephrine in patients with left ventricular failure secondary to long-term pressure or volume overload. Circulation 1983; 6 8 : 241-244. Ross J, Sonnenblick EH, Kaiser GA, Frommer PL. Electroaugmentation of ventricular performance and oxygen consumption by repetitive application of paired electrical stimuli. Circ Res 1965; 16: 332-338. 272 Roth RH. Action of angiotensin on adrenergic nerve endings: enhancement of norepinephrine biosynthesis. Fedn Proc 1972; 31: 1358-1364. Rubin PC, Scott PJW, McLean K, Reid JL. noradrenaline release and clearance in relation to age and blood pressure in man. Eur J Clin Invest 1982; 12: 121-125. Rush RA, Geffen LB. Release and clearance of circulating dopamine beta hydroxylase. Circ Res 1972; 31: 444-452. Saar N, Gordon RD. Variability of plasma catecholaline levels: age, duration of posture and time of day. Br J Clin Pharmac 1979; 8 : 353-358. Safar M, Pendler jp, Weil B, et al. Etude hemodynamique de 1'hypertension arterielle labile. La Presse Medicale 1970; 78: 111-114. Safar ME, Weiss YA, London GM, Chan NP, Milliez PL. Differences in the mechanism of borderline and permanent hypertension. Postgrad Med J 1977; 53 (suppl 3): 35-39. Schanberg SM, Kirshner N. Serum dopamine beta hydroxylase activity as an indicator of sympathetic activity and primary hypertension. Biochem Pharmacol 1976; 25: 617-621. Schmiedeberg O, Koppe R. Das Muscarin, das giftige Alkaloid des Fliegenpilzes. Leipzig: FCW Vogel, 1869. Schmitt H, Schmitt H. Localisation of the hypotensive effect of 2-(2,6 -dichloralphenylamine)-2-imidozoline HC1 (ST155, Catapresan). Eur J Pharmacol 1968; 6 : 8-12. Schoemig A, Fischer S, Kurz T, Richardt G, Schoemig E. Nonexocytotic release of endogenous noradrenaline in the ischemic and anoxic rat heart: mechamism and metabolic requirements. Circ Res 1987; 60: 194-205. Schwarz F, Thorman J, Winkler B. Frequency potentiation postextrasystolic potentiation in patients with and without coronary arterial disease. Br Heart J 1975; 37: 514-519. Schwartz L, Sole MJ, Vaughan-Neil EF, Hussain NM. Catecholamines in coronary sinus and peripheral plasma during pacing-induced angina in man. Circulation 1979; 59: 37-43. Sever PS, Birch M, Osikowska B, Tunbridge RDG. Plasma noradrenaline in essential hypertension. Lancet 1977; 1: 1078. Sharman DF. The catabolism of catecholamines. Brit Med Bull 1973; 29: 110-115. 273 Shepherd JT. Reflex control of arterial blood pressure. Cardiovasc Res 1982; 16: 357-383. Silverberg AB, Shah SD, Haymond MW, Cryer PE. Norepinephrine; hormone and neurotransmitter in man. Am J Physiol 1978; 234: E252-256. Simonsen S, Kjekshus JK. The effect of free fatty acids on myocardial oxygen consumption during atrial pacing and catecholamine infusion in man. Circulation 1978; 58: 484-491. Smirk FH. The neurologically-maintained componentin hypertension. In: Hypertension XXVII: Hypertensive mechanisms. Reader R, Ed. Circ Res 1970; 26,27 (suppl II): 65-84. Smith AD. Mechanisms involved in the release of noradrenaline from sympathetic nerves. Brit Med Bull 1973; 29: 123-129. Smith AD. Boichemical studies of the mechanism of release. In: The release of catecholamines from adrenergic neurons, Ed. Paton DM. Pergamon Press, Oxford. 1979, ps 1-15. Starke K, Endo T, Taube HD. Relative pre- and post-synaptic potencies of alpha-adrenoceptor agonists in the rabbit pulmonary artery. Naunyn-Schmiedeberg's Archives of Pharmacology 1975; 291: 55. Starke K. regulation of noradrenaline release by presynaptic receptor systems. Rev Physiol Biochem Pharmacol 1977; 77: 1-124. Stephenson RK, Sole MK, Baines AD. Neural and extraneuronal production by rat kidneys. Am J Physiol 1982; 242: F261-269. Stjarne J. Role of prostaglandins and cyclic adenosine monophosphate in release. In: The release of catecholamines from sympathetic neurons. Paton DM, Ed. Pergamon Press, Oxford. 1979, ps 111-142. Stromblad BCR, Nickerson M. Accumulation of epinephrine and norepinephrine by some rat tissues. J Pharmac Exp Ther 1961; 134: 154-159. Suko J, Ueba Y, Chidsey CA. Intracellular calcium and myocardial contractility. Circ Res 1970; 24: 409-413. Sundlof G, Wallin BG. Effect of lower body negative pressure on human muscle nerve sympathetic activity. J Physiol (Lond) 1978; 278: 525. 274 Sung CS, Virendra SM, Carcia E, Decastro CM, Hall RJ. Is postextrasystolic potentiation dependent on Starlin's law? Circulation 1980; 62: 1032-1034. Tarazi RC, Dunstan HP. Beta-adrenergic blockade in hypertension. Am J Cardiol 1972; 29: 633-640. Thames MD. Effects of d- and 1-propranolol on the discharge of cardiac vagal C-fibres. Am J Physiol 1980, 238: H465-471. Thomas JA, Marks BH. Plasma norepinephrine in congestive heart failure. Am J Cardiol 1978; 41: 233-243. Thoren P. Characteristics of left ventricular receptors with non-medullated vagal afferents in cats. Circ Res 1977; 40: 415-420. Tibbs PA, Young B, Ziegler MG, McAllister RG Jr. Studies of exptl cervical cord transection. J Neuro Surg 1979; 50: 629. Trendelenburg U. Release induced by phenylethylamines. In: Paton DM, Ed. The release of catecholamines from adrenergic neurones. Oxford: Pergamon Press, 1979; 333-354. Trimarco B, Massimo V, Ricardelli B et al 1981. Baroreflex responsiveness in borderline hypertensives : a study with neostigmine. Cardiovasc Res 1981; 15: 436-442. T uttle RS, McCleary M A mechanism to explain the antihypertensive action of propranolol. J Pharm Exp Ther 1978; 207: 56-63. Ulrych M, Frohlich ED, Dunstan HP, Page IH. Immediate haemodynamic effects of beta-adrenergic blockage with propranolol in normotensive and hypertensive man. Circulation 1969; 37: 411-416. Van Baak MA, Kho TL, Thijssen H, Rahn KH. Effects of acute and long-term beta-adrenoceptor blockade with propranolol on haemodynamics, plasma catecholamines and renin in essential hypertension. Eur J Clin Pharmacol 1982; 23: 377-382. Van Den Bos GC. E lizinga G, Westerhof N, Noble NIM. Problems in the use of indices of myocardial contractility. Cardiovasc Res 1973; 7: 834-840. Vanhoutte PM. Inhibition of acetylcholine of adenergic neurotransmission in vascular smooth muscle. In: Physiology of smooth muscle. Bulbring E and Shuba MF, Eds. Raven Press, New York. 1976, ps 369-377. 275 Vanhoutte PM. Introductory remarks: Alpha- and beta-adrenergic receptors and the cardiovascular system. J Cardiovasc Pharmacol 1980; 3 (suppl 1): SI. Vanhoutte PM. Adrenergic neuroeffector interaction in the blood vessel wall. Fed Proc 1978; 37: 181-186. Vanhoutte PM, Coen EP, Deridder WJ, Verbeuren TJ. Evoked release of endogenous norepinephrine in the canine saphenous vein. Inhibition by acetylcholine. Circ Res 1979; 45: 608-614. Vincent HH, Man In't Veld AJ, Boosma F, Wenting GJ, Schalecamp MADH. Elevated plasma noradrenaline in response to beta-adrenoceptor stimulation in man. Br J Clin Pharmac 1982; 13: 717-721. Vitanen K, Jaenne J, Frick MH. Response of blood pressure and plasma noradrenaline to propranolol, metoprolol and clonidine during isometric and dynamic exercise in hypertensive patients. Eur J Clin Pharmacol 1979; 21: 275-279. Vogel JHK, Jacobowitz D, Chidsey CA. Distribution of norepinephrine in the failing bovine heart; correlation of chemical analysis and fluorescence microscopy. Circ Res 1969; 24: 71-80. von Euler US. A specific sympathomometic ergone in adrenergic nerve fibres (sympathin) and its relations to adrenaline and noradrenaline. Acta Physiol Scand 1946; 12: 73-97. Wallin BG, SundBof G, Eriksson BM, Dominiak P. Plasma noradrenaline correlates to sympathetic muscle nerve activity in normotensive man. Acta Physiol Scand 1981; 111: 69. Walsh RA, Crawford MH, O'Rourke RA. Relative sensitivity of echocardiography and systolic time intervals for assessing acute positive inotropic interventions in human subjects. Am Heart J 1982; 104: 1061-1070. Watson RDS, Hamilton CA, Jones DH, Reid JL, Stallard TJ, Littler WA. Sequential changes in plasma noradrenaline during bicycle exercise. Clin Sci 1980; 58: 37-43. Watson RDS, Hamilton CA, Reid JL, Littler WA. Changes in plasma norepinephrine, blood pressure and heart rate during physical activity in hypertensive man. Hypertension 1979; 1: 341. Watson RDS, Page AJF, Littler WA, Jones DF, Reid JL. Plasma noradrenaline concentrations at different vascular sites during and isometric and dynamic exercise. Clin Sci 1979; 57: 545. 276 Weinshilboum RM, SchorottHG, Raymond FA, Weidman WH, Elveback LR. Inheritance of very low serum dopamine beta hydroxylase activity. Am J Hum Genet 1975; 27: 573-585. Weissler AM, Harris WS, Schoenfeld CD. Systolic time intervals in heart failure in man. Circulation 1968; 37: 149. Wetterberg L, Aberg H, Ross SB. Plasma dopamine beta hydroxylase activity in hypertension and various neuropsychiatric disorders. Scand J Clin Lab Invest 1972; 30: 283-289. Winer N, Carter C. Effect of cold pressor stimulation on plasma norepinephrine, dopamine-B-hydroxylase, and renin activity. Life Sci 1977; 20: 887-893. Wing LMH, Reid JL, Hamilton CA, Sever P, Davies DS, Dollery CT. Effects of clonidine on biochemical indices of sympathetic function and plasma renin activity in normotensive man. Clin Sci Mol Med 1977; 53: 45-53. Winternitz SR, Katholi RE, Oparil S. Role of the renal sympathetic nerves in the development and maintaenance of hypertension in the spontaneously hypertensive rat. J Clin Invest 1980; 66:971-978. Worstman J, Stuart F, Cryer P. Adrenodullary response to maximal str e ss in humans. Am J Med 1984; 77: 779-784. Yamaguchi N, de Champlain J, Nadeau R. Correlation between the response of the heart to sympathetic stimulation and the release of endogenous catecholamines into the coronary sinus of the dog. Circ Res 1975; 36: 662-668. Yamaguchi N, de Champlain J, Nadeau RA. Regulation of norepinephrine release from cardiac sympathetic fibers in the dog by presynaptic alpha and beta receptors. Circ Res 1977; 41: 108-117. Yamaguchi I, Kopin IJ. Plasma catecholam ines and blood pressure responses to sympathetic stimulation in pithed ra te s. Am J P hysiol 1979; 237: H305-314. Yamaguchi I, Kopin IJ. Differential inhibition of alpha-1 and alpha- 2 adrenoceptor mediated pressor responses in pithed rats. J Pharm exp Ther 1980; 214: 275-281. Yamori Y, Ooshima A, Okamoto K. D eviation of cen tral norepinephrine metabolism in hypertensive rats. Jpn Circ J 1973; 37: 1235-1241. Zenner RJ, de Decker DE, Clement DL. Blood-pressure response to swimming in ic e -c o ld water. Lancet 1980; i: 1 2 0 -1 2 1 . 277 $ Journal of Cardiovascular Pharmacology 6:954-960 S 1984 Raven Press. New York Changes in Blood Pressure and Plasma Catecholamines Caused by Tyramine and Cold Exposure A nthony J. I. Scriven, M orris J. Brow n, M ichael B. M urphy, and Colin T. Dollery Department of Clinical Pharmacology, Royal Postgraduate Medical School. London. England Summary: Tyramine may be used to stimulate release of 8 to 116 ± 9 mm Hg) and diastolic BP (from 72 ± 4 to endogenous norepinephrine. We have compared the in 81 ± 6 mm Hg). This was associated with a greater rise creases in blood pressure and plasma catecholamines in in plasma norepinephrine, from 0.357 ± 0.131 to 1.143 normal volunteers during (a) tyramine infusions, and (b) ± 0.393 ng/ml; plasma epinephrine was again unchanged. the more physiological sympathetic stimulus of cold ex A single oral dose of propranolol 160 mg caused approx posure. In a second study, the cardiac component of the imately a two-fold right shift in the systolic BP dose re pressor effect of tyramine was assessed by measuring sponse to tyramine. and blocked the tyramine-induced systolic time intervals, and by infusing tyramine in (3- shortening of the presystolic ejection period. Tyramine blocked subjects. Tyramine. 15.0 p.g/kg/min for 30 min. appears to exert its pressor effect mainly by stimulation elevated systolic BP from 122 ± 11 to 149 ± 4 mm Hg, of cardiac (3-receptors. This may account for the rela without increasing diastolic BP or heart rate. Plasma nor tively small rise in plasma norepinephrine (relative to cold epinephrine rose from 0.547 ± 0.184 to 0.836 ± 0.096 exposure) since the heart does not contribute a high pro ng/ml; plasma epinephrine was unchanged. Thirtv-min portion of circulating norepinephrine. Key Words: Ty exposure to 4°C elevated both systolic BP (from 105 ± ramine— Norepinephrine— (3-Blockade— Cold. The estimation of sympathetic nervous activity in study, incremental infusions of exogenous L-norepi- man is indirect and often depends on measurement nephrine also produced closely-correlated increases of the neurotransmitter norepinephrine (NE). It has in BP and plasma NE. However, the slopes of the been amply demonstrated that plasma NE concen regression lines relating BP to plasma NE were trations rise during sympathetic nervous system markedly different for the two amines such that stimulation by postural change, dynamic exercise, with exogenous NE, a 10 mm Hg rise in systolic and by physical and psychological stress (1-5); and BP was associated with a 1.0 ng/ml increase in conversely that attenuation of sympathetic nerve plasma NE, whereas for endogenous (tyramine-re- firing by sedation, ganglion blockade, and centrally leased) NE, plasma NE levels rose by only 0.12 ng/ acting sympatholytic agents is accompanied by a ml for the same BP increment. reduction in plasma NE levels (6,7). Tyramine releases NE from sympathetic neurons Such studies, while justifying the qualitative use although the mechanism differs from calcium-de of plasma NE as an index of sympathetic activity, pendent exocytosis (9-11). The aim of the present do not establish the sensitivity or quantitative va study was therefore to determine the relationship lidity of equating changes in sympathetic activity between BP and plasma NE when released on the and plasma NE in man. In a previous study (8) we one hand by tyramine, and on the other hand during showed that the release of endogenously-stored NE physiological stimulation of sympathetic activity by by the indirect sympathomimetic amine tyramine cold stress. Our previous study also suggested in raised both blood pressure (BP) and plasma NE; directly that the pressor effect of tyramine may be and when the increases in systolic BP were plotted partially mediated by stimulation of cardiac con against the increases in plasma NE, a highly signif tractility. This possibility has now been investigated icant linear relationship was evident. In the same in a separate study in which systolic time intervals Received March 2, 1984: revision accepted May 18. 1984. at Department of Clinical Pharmacology, Royal Postgraduate Address correspondence and reprint requests to Dr. Brown Medical School, Ducane Road. London W12 OHS. England. M 954 TYRAMINE, BLOOD PRESSURE, AND PLASMA CATECHOLAMINES 955 were used as a noninvasive index of cardiac con 10.0, and 12.5 p.g kg"'/min of tyramine), or in a higher tractility and the effect of propranolol on the dose range if they had taken active propranolol (5.0, 10.0, pressor response to tyramine was determined. 15.0, 20.0. and 25.0 |ig kg~'/min). BP, HR, STIS. and blood samples were obtained after METHODS 15 min of each infusion (our previous study with tyramine indicated that the increase in BP and plasma NE reached Tyramine and cold stress a plateau after 10-12 min). Six healthy male volunteers (mean age 27 ± 4 years; Measurements 66 mean weight ± 4 kg) had tyramine infusions or cold In all studies. BP was measured in triplicate by a semi stress studies 2 weeks apart and in balanced order. To automatic sphygmomanometer (Roche Arteriosonde). bacco. alcohol, and caffeine were avoided for 12 h be forehand. During all tyramine infusions the ECG was continuously monitored, and the heart rate was measured from the Tyramine infusions took place in a clinical laboratory ECG recording. In the cold stress studies, the heart rate and began at 0900 h with the subjects supine throughout. was obtained by counting the radial pulse. The amine was infused through a 5-cm fine gauge in dwelling polythene cannula inserted into an antecubital Systolic time intervals were recorded on Cambridge vein under local anaesthesia, and blood samples were equipment at a paper speed of 100 mm/s, using conven tional techniques to obtain total electromechanical sys drawn from an antecubital vein in the contralateral arm tole (QS:) and left ventricular ejection time (LVET) (12). via a 19-gauge butterfly needle, which was kept patent The pre-ejection period (PEP) was derived by subtracting with a slow infusion of heparinised 0.9% NaCl. the left ventricular ejection time from total electrome After 40 min supine rest, three sets of basal observa chanical systole, and corrected for heart rate (PEPI) by tions (BP. heart rate, and blood sampling for NE and EPI) the method of Weissler et al. (13). All intervals were cal were made at 5-min intervals, the mean results being culated from the mean of measurements on 10 consecu taken as the control. The subjects then received three tive beats, each read blind to the nearest 2 ms. incremental 30-min infusions of tyramine at 5.0, 10.0, and 15.0 gg kg-'/min: BP and HR were recorded and blood Blood samples for estimation of NE and EPI were kept on ice (5-10 min) until centrifugation at 3000 rpm for 10 samples were taken at 10. 20, and 30 min during each min at 4°C. Plasma was stored at -80°C until assayed by dose. a double-isotope enzymatic technique (14). Cold stress studies began at 0900 h and took place in The study protocol was approved by the Research a laboratory cold room maintained at 4-53C. The subjects Ethics Committee of the Royal Postgraduate Medical lay on a mattress placed on the floor and were covered School and subjects gave informed consent. Statistical warmly with blankets. Blood samples were drawn via an analysis was carried out using the paired Student's t test indwelling 19-gauge butterfly needle inserted under local and linear regression analysis by the method of least- anaesthesia into an antecubital vein, and kept patent with squares. Multiple t tests were not performed unless two- heparinised 0.9% NaCl. way analysis of variance showed a significant effect of After 40 min warm supine rest. BP, HR. and blood dose (p < 0.05). samples for NE and EPI were taken twice. 5 min apart, and the mean of these results was taken as the control. The subjects, who wore only shorts or lightweight trou RESULTS sers. were then abruptly exposed to the cold air temper Tyramine and cold stress ature by removal of their blankets. Cold stress continued The effects of tyramine infusions on BP, heart for 30 min: BP. HR, and blood samples were taken at 5. 10, 15, 20. 25, and 30 min. The subjects were closely rate, and plasma catecholamines are shown in Table observed throughout: there was no overt shivering al I. Systolic BP rose gradually with increasing tyra though superficial fasciculation was observed in some. mine dose, and this was accompanied by a small but significant fall in the heart rate. Diastolic BP Tyramine and propranolol was unchanged. Plasma NE levels also rose grad Five healthy male volunteers took part in this study, ually with increasing tyramine dose. It should be including three of the subjects from the first study. The mean age was 28 ± 3 years, and the mean weight 69 ± noted that the mean control plasma N E level of 6 kg. Intravenous cannulae for drug infusion and blood 0.547 ng/ml is higher than the value of 0.3-0.4 ng/ sampling were placed as already described. ml usually recorded in this laboratory for normal This study was conducted single-blind. Subjects at subjects at rest. This is because two of the subjects tended on two occasions. 1 week apart and received ei had high-normal resting plasma NE values (of 0.690 ther oral propranolol 160 mg or identical placebo tablets and 0.779 ng/ml, respectively) during both the ty on arrival in the clinical laboratory. ramine infusion and the cold stress study, although After an interval of 30 min (required for placement of their responses were similar to those of the other i.v. cannulae and monitoring equipment) the subjects subjects. Plasma EPI levels did not increase during rested supine and were undisturbed for a further 50 min. the infusions. At the end of this time, BP, HR, systolic time intervals (STIS), and blood samples were obtained twice at an in The effects of cold stress on BP, heart rate, and terval of 10 min, the mean of the two recordings being plasma catecholamines are shown in Table 2. Sys taken as the control. The subjects then received five in tolic BP rose significantly (although less than during cremental 20-min infusions of tyramine, either in a lower the tyramine infusions); the most rapid increase oc dose range if they had taken the placebo (2.5, 5.0, 7.5, curred during the first 15 min of cold stress, with J Cardiovasc Pharmacol, Vol. 6, No. 5, 1984 f 956 A. J. /. SCRIVEN ET AL. TABLE 1. E f f e c t s o f tyramine on blood pressure, heart rate, and plasma catecholamines (M e a n ± SD) Tyramine 5.0 gg kg'1 min 1 Tyramine 10.0 gg kg’1 min 1 Tyramine 15.0 gc kg 1 min 1 Control 10 min 20 min 30 min 10 min 20 min 30 min 10 min 20 min 30 min Systolic BP (mm Hg) 112 11 111=8 111=6 115 = 5 128 = 6" 130 = 5h 133 = 5' 145 = 8' 146 = 5' 149 = 4' Diastolic BP (mm Hg) 68 = S 69 = 6 68 = 6 68 = 8 71 = 11 68 = 7 72 = 5 69 = 7 69 = 9 72 = 10 Heart rate (beats/ min) 58 r 4 56 = 5" 58 = 4 58 = 4 55 i 4'’ 54 = 3" 53 = 5‘ 53 = 6' 53 = 5" 54 = 7h Plasma norepi nephrine (ptz/ml) 547 184 596 = 163 597 = 128 681 = 192 739 = 154'' 826 = 140' 747 = 150“ 778 = 115' 959 = 250' 836 = 96' Plasma epi- nephrine (pg/ml) 89 60 62 = 44 51 = 36 70 = 53 57= 29 58 = 33 63 = 25 62 = 50 122 = 62 84 = 51 " p < 0.05, '’p < 0.02. 'P< 0.01 versus control. no further increase in the last 15 min. Similarly, Effects of propranolol on the response plasma NE levels rose during the first 20 min of to tyramine cold stress and were essentially unchanged during The effects of 15-min incremental tyramine infu the final 10 min. Neither heart rate nor plasma EPI sions on blood pressure, heart rate, and plasma NE levels changed but in contrast with the tyramine after placebo (control) are shown in Table 3 (con infusions, diastolic BP rose progressively and sig trol); and after oral propranolol 160 mg in Table 4. nificantly during cold stress. The only change induced by propranolol in the In Fig. 1, the BP and plasma NE data from Tables baseline (pre-infusion) data was a small increase in 1 and 2 have been plotted to show the relationships resting diastolic pressure (p < 0.05), although sys between mean plasma NE concentrations and mean tolic pressure and heart rate were slightly but not increases in systolic pressure obtained during ty significantly reduced. Propranolol did not alter ramine infusions and cold stress. This shows that resting plasma NE levels (compared with placebo), plasma NE is closely correlated with increases in and plasma NE levels at the 5.0 and 10.0 p-g kg-1 systolic pressure during release of endogenous NE min “1 tyramine doses were nearly identical to those both by tyramine (p < 0.01) and by cold stress (p observed after placebo. < 0.01). It is also clear, however, that these rela Figure 3 is a log dose-response plot showing the tionships differ quantitatively: the slope of the rise in systolic pressure with increasing doses of regression line for tyramine infusions is 80, and this tyramine after placebo and propranolol. Both is markedly steeper than the slope of 16 obtained curves have similarly shaped configurations, but the for the cold stress studies. curve after propranolol is shifted to the right in an Figure 2 shows the relationships between in approximately parallel fashion. When the data from creases in plasma NE and increases in diastolic the straight portion (i.e., the upper four doses) of blood pressure. Tyramine infusions were not asso each curve were replotted to show the relationship ciated with increased diastolic pressure, and there between A systolic BP and the corresponding was no relationship between the plasma NE level plasma NE level, it was evident that propranolol and diastolic pressure changes. By contrast, the in caused a parallel shift of the A systolic BP/plasrna creases in diastolic BP during cold stress were NE curve to the right (Fig. 4). closely correlated with the plasma NE and the slope The resting PEPI was 118 ± 6 ms after placebo, of the regression line (13) was similar to that ob- and 126 ± 7 ms after propranolol, but this differ , tained for systolic pressure. ence was not significant. As shown in Fig. 5, tyra- TABLE 2. E f f e c t s o f cold stress on blood pressure, heart rate, a n d p lasma catecholamines Cold stress Control 5 min 10 min 15 min 20 min 25 min 30 min Systolic BP (mm Hg) 105 = 8 111 * 10“ 114 i 9 b 115 ± 10f 117 ± IP 116 ± 9C 116 — 9° Diastolic BP (mm Hg) 72 ± 4 76 ± 5“ 77 ± 8 79 ± 5‘' 79 * 5“ 80 = 5C 81 ± 6C Heart rate (mm Hg) 55 * 6 57 ± 9 55 i 8 55 r 9 54 ± 10 55-9 55 ± 10 Plasma NE (pg/ml) 557 ± 131 744 = 213* 790 ± 236“ 892 4: 251“ 1069 ± 432“ 979 = 330“ 1143 ± 393' If Plasma E (pg/ml) 68 — 37 O 54 ± 32 54 ± 34 67 ± 45 61 ± 43 64 ± 37 “ p < 0.05, bp < 0.02. cp < 0.01. pl,„r~,„rnl V„1 f, \1„ * 10XJ TYRAMINE, BLOOD PRESSURE , AND PLASMA CATECHOLAMINES 957 40- 30 0L J 20 j FIG. 1. The relationship between 0 plasma norepinephrine and increases >-n in systolic BP during tyramine infu /) sions (closed circles) and cold stress z 10 (open triangles) in normal volunteers. LU -10- j______i______i______i______■ _l______L 0-5 0 6 0-7 0-8 0 9 10 12 PLASMA NOREPINEPHRINE (ng/ml mine produced a dose-related shortening of the been examined in this study by comparing the ef PEPI (after placebo); this effect was abolished by fects of tyramine and cold stress. The technique of propranolol. prolonged cold stress reported here has not been previously described as a deliberate method for DISCUSSION stimulation of sympathetic nervous activity, even The relationship between changes in BP and in though various forms of exposure to cold are known plasma NE following endogenous NE release have to elevate both BP and plasma NE in animals and 10- xa> /A E E A A A COLD STRESS Slope = 13 FIG. 2. The relationship between A r = 0-95 plasma norepinephrine and increases p < 0 001 in diastolic BP during tyramine infu sions (closed circles) and cold stress (open triangles) in normal volunteers. TYRAMINE Slope = 0-2 r = 0-17 N.S. j______I______I______I______I------1— _L_ 05 0-6 0-7 0-8 0-9 10 1-2 PLASMA NOREPINEPHRINE (ng/ml) J Cardiovasc Pharmacol, Vol. 6, No. 5, 1984 958 A. J. I. SC RIVEN ETAL. TABLE 3. Effects of tyramine after placebo Tyramine dose (|xg kg 1 min ~') Control 2.5 5.0 7.5 10.0 12.5 Systolic BP (mm Hg) 114 ± 5 115 - 10 116-8 121 - 9" 132 = 4r 138 - 6‘ Diastolic BP (mm Hg) 63 - 4 62 ± 5 64 ± 4 63-5 62 ± 7 62 = 3 Heart rate (beats/min) 64 - 9 62-7 61-6 60 r 7 60 ± 10 60 - 10 Plasma NE (pg/ml) 283 ± 62 285 ± 70 350 ± 57 419 r: 44<' 445 - 103* 461 r 96‘ " p < 0.05. bp < 0.02. ‘p < 0.01. man (15-18). In many ways, 30-min exposure to has been found to be a large contributor to circu cold air is a very different form of stimulus (of sym lating NE (23); and plasma NE concentrations cor pathetic activity) from the infusion of tyramine; yet relate well with direct recordings of muscle nerve it is clear that the two stimuli are comparable in activity (24,25). several important respects. First, subjects are It is evident that BP responses during the two studied in a supine position throughout, thereby procedures differed in type as well as in magnitude. avoiding postural changes and their associated hae Thus during cold stress, both diastolic and systolic modynamic and sympathetic effects. Second, the BP increased equally with no widening of the pulse period of observation in both cases allowed suffi pressure. This cold-induced pattern of haemody cient time for NE concentrations to achieve a new namic change is consistent with a dominant in steady-state in plasma (19) — in contrast to brief crease in peripheral vascular resistance due to ar stimuli such as the cold pressor test (1,15). Third, teriolar constriction, with little or no change in car tyramine and cold stress appear to be specific diac output. Previous studies have indeed stimuli of NE release, with no increase in plasma demonstrated that changes in cardiac output during EPI levels which might complicate the interpreta exposure to cold are negligible (21). With tyramine. tion of blood pressure changes. As these conditions however, a different response was obtained. Dia would not have been met by other commonly used stolic BP was unchanged, but the systolic BP in experimental techniques for the stimulation of sym creased and the pulse pressure widened suggesting pathetic nervous activity, it seemed appropriate to that tyramine raised BP partly by its stimulation of use the method of 30 min cold exposure described cardiac contractility. This view is supported by the above. This was confirmed in preliminary experi result of the second experiment. Here, propranolol ments, and borne out by the results presented in induced a rightward and approximately parallel this paper. shift not only in the tyramine dose-response curve, In normal subjects, cold stress induced a modest but also in the plasma NE/systolic BP relationship. but significant elevation of both systolic and dia Moreover, the tyramine-induced abbreviation of the stolic blood pressure of about 10 mm Hg. Similar PEP was abolished by propranolol. This indicates BP increases have been reported by other groups that part of the pressor effects of tyramine is sec who have studied the haemodynamic effects of pro ondary to stimulation of (3-adrenoceptor mediated longed cold exposure (4,20), although in older sub cardiac contractility, since propranolol attenuated jects the rise in BP can be as much as 40 mm Hg both the cardiac inotropic effects and the rise in BP. (22). Subjects felt that the stress of prolonged cold Despite the blockade by propranolol of vasodilator exposure was not just a pain stimulus such as that (32 adrenoceptors in peripheral vessels (thus un experienced during the 2-min immersion of the hand masking a-mediated vasoconstriction) tyramine in iced water employed in the cold pressor test. failed to elevate diastolic BP except at doses higher Although overt shivering was not observed, the than those used with placebo. Clearly, however, ty ' large increases in plasma NE could still reflect a ramine caused some degree of peripheral arteriolar largely (skeletal) muscular origin since this tissue constriction since there was no change (or a small TABLE 4. Effects of tyramine after propranolol 160 mg Tyramine dose (p.g kg 1 min ') Control 5.0 10.0 15.0 20.0 25.0 Systolic BP (mm Hg) 111 - 8 112 = 8 114-9 120 ± IP 130 - 20" 133 ± 18* Diastolic BP (mm Hg) 68-6 69 ± 5 70 ± 8 74 ± 4“ 77 ± 9 82 - 9" Heart rate (beats/min) 60 ± 7 58 ± 8 56 ± 6 54 ± 5" 54 - 6b 56 ± 9 Plasma NE (pg/ml) 279 - 63 385 ± 67 449 - 89" 542 - 1031' 589 ± 201" 626 ± 118f " p < 0.05, *p < 0.02, ‘p < 0.01. J Cardiovasc Pharmacol, Vol. 6 No. 5, 1984 1 TYRAMINE, BLOOD PRESSURE, AND PLASMA CATECHOLAMINES 959 30- —o> 25 X E JE 20 u j 2 0 ITZ3 cn cn UJ ^ 10 £ 15 o (/) _l o z 01- c/5>- 10 LU co to z ai< a: UJ U-10 2Ul 5 oce TYRAMINE ( pg/kg/min ) z 0 FIG. 3. Effect of tyramine on systolic BP after placebo _]______I______1_____ I_____L (closed circles) and oral propranolol 160 mg (open circles) 0-3 04 0-5 0-6 0 7 in normal volunteers. Mean r: SEM. PLASMA NOREPINEPHRINE ( ng/ml ) FIG. 4. The relationship between plasma norepinephrine and fall) in heart rate despite the increase in cardiac NE increases in systolic BP during tyramine infusions, after pla release; so there was presumably a sufficient rise in cebo (closed circles) and oral propranolol 160 mg (open cir peripheral resistance to activate the baroreflex and cles), in normal volunteers. increase vagal drive to the heart. The heart rate anisms; the pressor effect is in each case accom response is therefore the net effect of tyramine's panied by an increase in circulating NE. However, cardiac action (due to NE release in the heart) and because the pressor effects of neuronal NE release of the increased vagal activity; some subjects show will depend on the function of the target tissue, a a slight bradycardia while in others there is no given rise in BP will not always be accompanied by change — unless the cardiac action is blocked by the same increment in plasma NE. This is evident propranolol, as in our second study. from the dissimilar plasma NE/BP relationships ob These differing patterns of haemodynamic re served for tyramine and cold stress, which dem sponse (to cold and tyramine) appear to reflect the onstrate that even with endogenous NE release, the varying contributions of the heart and of the resis plasma NE/BP relationship may vary even within tance vessels to the acute pressor response. Stim an individual. Furthermore, the relative contribu ulation of NE release in either of these “ sectors” tion of different tissues to circulating NE is not al raises the BP independently and by different mech- ways proportional to its sympathetic innervation or FIG. 5. Effect of tyramine on the pre-ejection pe riod index (PEPI) after placebo (closed circles) and oral propranolol 160 mg (open circles) in normal volunteers. J Cardiovasc Pharmacol, Vol. 6,5, No.1984 960 A. J. I. SC RIVEN ET AL. the amount of NE released during sympathetic ac 13. Weissler AM. Harris WS, Schoenfeld CD. Systolic time in tivation. and this may be due to the large intertissue tervals in heart failure in man. Circulation 1968:37:149 variability in synaptic cleft width with consequent 14. Brown MJ. Jenner DA. A novel, double-isotope technique variability in NE spillover (23,26). for enzymatic assay of plasma catecholamines, nermitting The interpretation of plasma NE levels has not high1981:61:591-6. precision, sensitivity and sample capacitv. Ct n Sci 15. LeBlanc J. Cote J, Jobin M. Labrie A. Plasma catechol always reflected the anatomical and functional com amines and cardiovascular responses to cold and mental ac plexity of the sympathetic nervous system and the tivity. J Appl Physiol 1979;47:1207-11. variable kinetics of circulating NE (27). There is 16. Winer N. Carter C. Effect of cold pressor stimulation on now an increasing weight of evidence from animal plasma norepinephrine, dopamine-fi-hydroxylase, and renin activity. Life Sci 1977:20:887-93. studies showing that sympathetic nervous system 17. Johnson DG. Hayward JS, Jacobs TP, Collis ML. Eckerson activation is not always generalised, but may be JD. Williams RH. Plasma norepinephrine and responses of directed selectively at specific organs in accordance man in cold water. J Appl Physiol 1977;43:216-2'.’ with physiological requirements (28,29). Although 18. Picotti GB. Carruba MO, Ravazzani C, Cesura .1. Galva the effects of tyramine are not physiological, this MD, Da Prada M. Plasma catecholamines in rat:-. exposed to cold. Eur J Pharmacol 1981:69:321-9. study demonstrates that in man also the release of 19. FitzGerald GA. Hossmann V, Hamilton CA, Reid JL. NE can have variable effects depending, it seems, Davies DS. Dollery CT. Interindividual variation in kinetics in which tissues the NE release occurs. A large con of infused norepinephrine. Clin Pharmacol Ther tribution to NE originates from tissue such as skel 1979:26:669-75. etal muscle (23), which is not directly involved with 20. Hattenhauer M. Neill WA. The effect of cold air inhalation resting cardiac and circulatory function, whereas, on angina pectoris and myocardial oxygen supply. Circula tion 1975:51:1053-8. the myocardial contribution to circulating NE 21. Epstein SE. Stampfer M. Beiser GD, Goldstein RE. Braun- seems relatively small (30-32), and even when in wald E. Effects of a reduction in environmental teir-erature creased, probably insufficient to cause significant on the circulatory response to exercise in man. .V Engl ] Med 1969:280:7-11. elevation of plasma NE levels (33). 22. Zenner RJ. de Decker DE. Clement DL. Blood-pressure response to swimming in ice-cold water. Lancet 1980: i: REFERENCES 120-1. 23. Brown MJ. Jenner DA. Allison DJ, Dollery CT. Variations 1. Robertson D, Johnson GA. Robertson RM. Neis AS. Shand in individual organ release of noradrenaline measured by an DG, Oates JA. Comparative assessment of stimuli that re improved radioenzymatic technique: limitations of periph lease neuronal and adrenomedullary catecholamines in man. eral venous measurements in the assessment of sympathetic Circulation 1979;59:637-43. activity. Clin Sci 1981;61:585-90. 2. Kozlowski S. Brzezinski A, Nazar K, Kowalski W. Fran- 24. Wallin BG. Delius W. Hagbarth K-E. Comparison of sym czyk M. Plasma catecholamines during sustained isometric pathetic nerve activity in normotensive and hype;;ensive exercise. Clin Sci Mol Med 1973;45:723-7. subjects. Circ Res 1973;33:9-21. 3. Watson RDS. Hamilton CA. Jones DH. Reid JL. Stallard 25. Wallin BG. Sundlof G. A quantitative study of muscle nerve TJ. Littler WA. Sequential changes in plasma noradrenaline sympathetic activity in resting normotensive and hyperten during bicycle exercise. Clin Sci 1980;58:37-43. sive activity. Hypertension 1971;1:67-77. 4. Cryer PE, Haymond MW, Santiago JV, Shah SD. Norepi 26. Bevan JA. Su C. Variation of intra- and perisynaptic adren nephrine and epinephrine release and adrenergic mediation ergic transmitter concentrations with width of synaptic cleft of smoking-associated haemodvnamic and metabolic events. in vascular tissue. J Pharmacol Exp Ther 1974;190:30-8. N Engl J Med 1976;295:573-7'. 27. Esler M. Assessment of sympathetic nervous function in 5. Lake CR. Ziegler MG. Kopin IJ. Use of plasma norepi humans from noradrenaline plasma kinetics. Clin Sci nephrine for evaluation of sympathetic neuronal function in 1982:62:247-54. man. Life Sci 1976:18:1315-26. Car- 6 . Wing LMH. Reid JL, Hamilton CA, Sever P. Davies DS. 28. Shepherd JT. Reflex control of arterial blood pressure. Dollery CT. Effects of clonidine on biochemical indices of diovasc Res 1982:16:357-83. sympathetic function and plasma renin activity in normoten- 29. Abboud FM. Integration of reflex responses in the control sive man. Clin Sci Mol Med 1977;53:45-53. of blood pressure and vascular resistance. Am J Cardiol 7. Murphy MB, Brown MJ, Causon RC, Scriven AJI, Dollery 1979:44:903-11. CT. The effects of pentolinium on sympathetic activity in 30. Yamaguchi N, De Champlain J, Nadeau RA. Correlation hypertensives and normotensive controls, bit J Clin Pharm between the response of the heart to sympathetic stimulation Res 1983;3:227-31. and the release of endogenous catecholamines into the cor 8. Scriven AJI. Dollery CT, Murphy MB. Macquin I, Brown onary sinus of the dog. Circ Res 1975:36:662-8. MJ. Blood pressure and plasma norepinephrine concentra 31. Cousineau D, Ferguson RJ, De Champlain J, Gauthier P, tions after endogenous norepinephrine release by tyramine. Cote P, Bourassa M. Catecholamines in coronary sinus Clin Pharmacol Ther 1982:33:710-6. during exercise in man before and after training. J Appl Phy~ 9. Burgen ASV. Iversen LL. Inhibition of noradrenaline up siol 1977:43:801-6. take by sympathomimetic amines in the rat isolated heart. 32. Haneda T, Miura Y. Arai T, Nakajima T, Miura T, Honna Br J Pharmacol 1965;25:34-49. T, Kobayashi K, Sakuma H, Acachi M, Miyazawa K, Yosh- 10. Smith AD. Mechanisms involved in the release of noradren inaga K, Takishima T. Norepinephrine levels in the coronary aline from sympathetic nerves. Br Med Bull 1973;29:123-9. sinus in pateints with cardiovascular diseases at rest an 11. Trendelenburg U. Release induced by phenylethylamines. during isometric handgrip exercise. Am Heart J 198 , In: Paton DM. ed. The release of catecholamines from ad 100:465-72. renergic neurones. Oxford: Pergamon Press, 1979:333-54. 33. Rose CP, Burgess JH, Cousineau D. Reduced aortocoronary, 12. Lewis RP. Rittgers SE, Forester WF, Boudoulas H. A crit sinus extraction of epinephrine in patients with left ventric ical review of the systolic time intervals. Circulation ular failure secondary to long-term pressure or volum 1977;56:146-58. overload. Circulation 1983:68:241-4. I Csir/iinvn tr Phnrrrulrnl Vr> I f, “i SjnIQXJ a a |P a s < a Sympathetic Nervous System Activation in Postextrasystolic ooooooooooooooo < < < < < S U < S S S < < U S Potentiation: Role of Catecholamine Release in Enhancement of S S c f S S j U S ^ c f S S S S ■ S ■3 iS fa #o ■S a> c u u 3 s u c j tu uj y*z O tU Q£ z o < 5 < Q Q - < 5- s S S S S < U S < S < S < S < U *, * ‘S', *T \D v*', -*T ^ oociooooooooooo i r i ^ ^ ' S v > — •£>•£ S- -c 'S'- j;a. ^ I 1 I .1 1a£ t ii£ t tag HS jo ; 5i jto— ';: <= n to S 2 2 3 ■£ w — "Ht* 5 £ •—CJ ~■| -8 «. E . ?« .2E c .E« to;B Z uJ3 e 8 2= §i o J u .£c , E Eos 5efl r - U *- 3o J* > o g»| a 2O * to 3 .1S’ i O .e -i ; -u r; ?c S. tp3 S 2 £.2 E ^ 2 5 . u . s « -g 'E .E& c ■= — = .2?c S E S -a C r T3 j: S ^ g H OCU 2 u ca- ca- — o~ S u * Eu* ;f 3 'C JC“ H .5 2 5 & 5 JC ^ c 5 wO 0 «8. o— O o— o c 2 go ow is OCC w K - - ^ c £ 2 '5 -2o&, 2 ■o^ o o u c to: o ^ c ^ S£ - = 5 S to “ g £ is « E 2 0 £ d 9 . *E w 2 _ _ _e 2 w — £ g £ S oo t uuo 2 ^ 22“9 E . ^ - - g> - - 5 1 toS ■ 1 N Q U Q. N 15»■ C 5 8 t oo o « > cl g ~ .2 *s“ CJ s -6 ^ w> _ 8 o s ■» ■ *3 o 8. § "8 o — “ .§ S-c-c •aE *3 .o ■a 0 8-i " t»5E_ HS l*'S5 c >>I.H6 u o ° ~ S-a " n.=j \s O£C i «ua ii =c ca a . f i g E g a. i f O. g£3 q ^ O M'S.ii aj « > > £ ' S a S £ .1 S T3D. C« 0^ ■£ i£ 3 £ ?= .2o. <« 2 3 £ 3 E c ^ g g o c s-s gs-sy e jc .E | « e 2 w2 =5>> fa “ ■— ? OJ F-= CO w * 5 U - 3 t 3 .« ft » cc .s •=—E C 8.1 o S V .C*O co _ - E - H . 2 > , 2 J i c X 2 X c E g i = 5 DO ^ u U 5 3 « T3 .5 p s w cs rp B8 S3g 2 ^ g *o ■E .1 -a to !3 T3 i E i t C E ■s“> «£-oi O .p O 3 .2c -2 t>> £ ^ tw R, JS ^ L- 5 e-rci DO «g 2 E2 ,o w>w tS >> ^ •- 8o p e3 -E o S n >> b T5 i o> !£ TZ . S C w g b i ) w w aj CL t: o IS ^ C^ 8. o c .5 o uCu.o 3 3 a -ri J£ - •aS = .yc; -15 2 ■ 1 * S «£ .2 “ = o= •; ^ SS < ■£ a see ts 00.2 * CO 2E i. « u < •£ T3 8S..C 0 c o a •§a -a 2 8.1£ S» - w _C 3.. g> s " 8 " •sS f •3 H w £ 52 -C JD fS2 a *- a. u S DO ^ A 00^ c — s .O « 6 U -Co oc 8*1 d 4 C ra JE * “O *3 w -c £ B +l -T3 o o 3 DO t2 >. E c g c oo C' 3 * 4> E — o < jc "82 g*'-g §o d “ E > +1 n ch . .EQ. £ £ V 2 00E 'w 2 £ 8 '§*" 1 CuUb-S- 3 J C S « o S P 2 5 U cn §e = :§ £ 30. “ s € g. y So. 22uE oo > “ -SK . | i 8!! ‘ i c _* | -s. « < i§ .2c|JS -ac ■ 5>«££ •O a|EE|S JJ-=U T1W E S n u £ * « 4>a. .2 o — to 18 ? « CO‘C u v5 a■ c rz uw a o> oI t- JCc** s t s eo i<§ 'J= I I -32 £3* E 00O 2 2 | ,g>2| 2 c ■8 .■ ?3 SS' g>£g 18 BsS -tl s a «■•= 'E if 5«>; U 2-Si 18 £ 1 ^ *5 flj W to:> 2 3 , 3 8 g c sz T3 - ; - •I| 'fi 111 ^ o S o ^ » •nU * X un ai 2 c * .2 w O*, tn p. Q. cw» t>C C8w * *0 <3 _ E -a 2 ;-ssSs : | 4 I U c g c t u a t on ; r|l'5 R 2 2 !Sg .2§ y | is ^-.2 " ’°2 j . c ^ u / ^ ^ Cfl^ C3 U«- 3 eo ^ >>"o E T3 —• o * ^ ■? aa 0 i ”'2 !/^11 -O u.-o it o < % V- Jg J E u l i GESCHW1ND tl al JACC Vol. 4, No. 2 CATECHOLAMINES IN POTENTIATED BEATS is X u 2 Q < l t % i August 1984:215-25 z 2 u h- 1 < tl S»»© ■ .s 5 • S >. E ■5 i i E a ^ i i — 2— “ Vo .S-3 !■-i-il i - i - £!§■ 5 Si H -a u | e - »«§ e - u| O >< a>; 2 o o «o o o: c y ft> C. ^ o O •— c e -o a*3_ t:U o •- c c o>o ~ £ " e— g ' P - b _o X + ' + ' 5 s s O •S =5 lu 2 E o p | f P i . Q u ^ <; •- “ £ l % u 'a. fa *-» o § » c4 _ S>5 q c b 5 C C« o o o 3 iH ^ T T -ft r^, ^ fN) 33 « .£ i2 O)i sO 1L 1 o0 o 5 ©— o >: o o > fc w 1/5 c c \ / - o r a £7OX) o c |£ £ | ^ ■B ~ o _ ■5-^2 ■5-^2 - x> g So - ? 2 o e *5 °*2 3 ^ -S5 ^ a o *P — O C O « 5 E » .£ 50 W.S- o S y & y 4= O d * _ r* _ id *3 i § s I o ‘.5 5 « ~ V5 « 3S *E U § - S i ,-c 5 O 2 c a, O u OX) * ^ is ^ -5 y “T a o 'Cw c *3E a o K c g S J s e s > . S3 iS » M 3 > 2 « i ; « c 3 • «« C ■ ! * !■“ “ » ^ ^ E ■S t O Z, £ -e )3 o o . rj* ed £ 2 Jc .5 £ „ .Sg o « | 1 1 ^ ^ _ ^ 5 "5^ •& ‘E. +1 J=c b ^ 0 o o '= : -P so H 011 £ Q5 n o t-* +1 83 — E ■D +l S 0>S$ -n V a. — 03 a. y _i 0.0 "S a. M o 2 3 c.3 d 3 b X ^ to c« / s • ^ S o y w £ U N 3 .E c ? rs.il .r-s c E^ O g *rr^ ^ 0) g •- 3 s 00 00 L» » C CI E 3 “2 60 Oil b g b .s l.E | |s C 3 O "*£ o J= 2 00-2 , ca .c ^ ■g s ^ s ^ ■g s .. £? j.^ 6 + ■2 = 8-Z I -r. *5 J - c _ cl ® p r aS- o o w b ™ o « z ! I s s I ! S ; — . So :t c > - -D ; " 5: b» fc- . S 5 c C § o o c > c C/3 2 o o n. rr o in ° J 1 « z M .2 05 O. *5 p IK X T c ^ £ J5 “ 5 a a F u ^ 2 3 • M .5 g §8. a _ _ s S V “ o O •E "3 ™ l g.'*-u » c > 8 5 b 6 c s O ‘3 2 c ^ ^ IIS'S, «J . _ ==I | g j>b £> *"*— 2 . j 3 j J= g'Z -a ~ «d = - 3 3 "3 a ■-~a E 5 o > ~ tj j ^U2 'C e d *JS — Jd •- 2 4J ed o cd 0 : I O 3 C 4J 00 > *d s “ a s •s ^o s g ,o -u *5 C 3 ^ r- n § £ ««£■ §■ I -§ 6C 0 O 3 O • £ O. a u « O - * u £ 5 ~ = s i ) ^ - 2 c o ^ 'OQw w Q u o O Co ' > c O O 5 W OX) fc- uj uv . ra 3 .x u v. • In ^ ^ ^ C C j C ^ C — ^ 1 + < 3 *°.£ co^0> •2 c u oo'u60 y (j j n 8 >» „ r -a = " ' I I 'i= 4 60 O o . £ — T “ S , 2 P c L ' = c o -a 0 0 g* - -I 1 « 1 ^ I •— "5 St tji — T O ~ 2 c § •o — T 3 "T3 Q . ^ *“ o» * o 0 i §■5 | a • c *5 !£' *5 3H .. p ^ ! r E s j : «— o z f : t 3 8M-i5' -c '2 5 i - ‘8 M £ c ° * c w o ' S “ E | «E S ' ! “ | — — .E co S - " 3 O ! s o u u-O CO o ? I 8,6gI I - s t l i l k c W , 5 - = o c E c E E cd 3 3 “ 8 C 3,. J = ^ 5 „ g l^'g c c _ _ '— _ c c *- u- ^ Is I t £ c — b o u a .3 y '•/ £ .2 3 z Ji -= .2 -c c 2 cO' r,o'—u - p e c .Si.■B55 c ca 5 “ H a .E s s d ) 5 u a b < P • = O 6 f t ^ ^ u] • S 2 . • £ . 1 = £ B Z Z g o - a - o g ^ I ■£ § ‘ 3 * 1 1 * 8 3 »;3 «.U E •§ S.S®> ® S -S . > ) 0 U 3 - © -a o *S o ai *T3 *b D.# n » 1 2 t § » ^ ’O 2 £ . 2 ■- E < . Q 3 _«5 -— .5 | o s'o-^'SS c3 2 " 3 < — § E & £ E tj ^ Ci» «*. SJ8 g i> «3 o5 y +1o U ^ -O oS C o . «d J 4> -C 0 I I 8 I Vrt « 3 c^ « | | Cj — • S -3 -r }a cu’H m ?vm E c - Q.*- O {J O o * C ’ 3 C 9 +12 § -c is E 5 .; a 0 Jr c u g fe g--E O i H ^ U §£ C c- C ^ t> 2 g U S fc-u c ^ -j-. 9 > U9 o w c o o c^ f: o « >* V5 w 3 c 3 is 5 -a T, ra .5 Z cd C 3 c ■“ -c .£ o ^ g « •- o ° o n ® i _ - 2 if X 0 ^ 2 ^ V g= => 0 3_ ^ a . g s , S + 1 • 5 - S ^ . U + 1 i.s 3 W)r o E >. ■°K B O - — O — 3 C ~ x 2 k ^ g uC 3 W 5f c . C S C l t - ^ . l l | i 00 S O ■ > y “ >. ”ra •O O . , 'C ^ 5 u u o o . u T) 35 -o M « ■5 Cd «00 s .E B >►•= x:w £ — •S J3 u g Cu M gfc; a .sp.'s I S -E£ I g ^ ? & *c D C/2c -2 2 ” 5 Cu 0 3 1 o -s 0 O c H ^ 3 E 3 r s 2 § B ' a r t g u „ "> « , § , o i i C * c^ S.8 a i c +1 2 so ?3 CO £ O. id W w r3 t5 ■£ d Vg T3 > c. ^ 1 ^ ^ §1 “ 1 g 1 £&§•+< 2 . Jg - c ^ ^ * c -3 — >>— c ■ ‘Csd o> £ - ^ r ^ - , s \n - — *o I ■ = ^ ; r3 c >;£ ^ — C o -3 £ UJ t "c U u. ffl y 3 K 3 S c £ — — — 0) X JJ u 'C c - £ £ Hr -5 _ s_ *° cd ■ Jb cd o 1 o- B c o -g 00 'C o c H x ^3 ?-3 H h cd — w Tl O) D. Vi c C 5 3 « ^ p . > .. 00 00 >, t) ^ ^ .£ «£^ 3 s- ^ -2 j3 i: >\ r •c ••' C o'*o u o g - 5 -3«C 4i £ — o 3 £ £ ^ a | u £■'£.—u I ^ I . 5 ^ |£ l o E E — 3 cd v S*^w-)0»n0^0 » £ ^ a» s -S g c s b b s u ^ S “ = g.-g 7 m g I IS ;2 - Sp ^ - O-a Is .2 .£ u£i o ao O o «■y; a ”CL-C d 8 o 4) «-> c c u o c « 0 — CO) ^5 =0 .^ U ' j -~ 0 < S S q inus catt ■‘ “ ~ V o - -s: s c •S 2 — uj r o o r , a - . — u f v i = 0 *£ + 1 ■o p s 1 £ = S= » E -g o C o 00 , x s £ - n y o _ E 3 n 3 b ■v. This Si -2 ' *3o' • — T Sf*T >> 0 S E 3 .3 rM o o rsi E w +1 3 P3 0 a.H 2 „ x>d flow '0 ej -8 a g. o ^ S 2 r H « ' — i ^ - = w , e 2 ■ ■ 5 .*5 O 3 1 -5 -3 o Ko > 2 ned by: 5 o ■2 c£ 8 W£ O 0 5 O M)*“ g ^o"c cL H ? >t « b .■o .' v S i 2 e p 00 f—J.i ts .£ .1cd” Q. o C 60 •C ~ --« E E ? — ■3 cd »c y «c 1 S.E ^|is w c £ Q. y o ^ •£ I<1 U. C .2f o cj .0 « J -o il C O > c o c e 3 « l $ 2 i « o •a 5! ■3 b S is •= ■- g S O c > ^ b5r o © .J! $ j < 5g g ^ ■§ § ■§ S.M ^ «^ C ^ •£ & p « > j I Q s Q •c j3 -a E o 55 8 g y 5b ■a §. i ~ C 2 10 51 21 c 00 s 55 .5 ed 3 C ~ 00 2 -fi gsJ{|- -fi 2 r a ^ . 5 r a^ S - ^ * 3 r - O *^O . 5 C : © l o - I p-.g = i> I w .5 03 3 JJ I'f X I i 8 73 i g I ! E 23 X w p 1= j j ?3 I H M “3 U CW c. c c •C o E f t - - 3 - C - -c 3 >, -s c _c c *— w: C ‘g* « o g SB •n § “ o o o 'a- u c - B O u -o O) w p J. Q. w C w .3 | | o — D.— (-9 J °°£ J ??— C ‘S. .2 E E E ao -c £ £ O- o s ^ -- X b c b i j C > <^ cd o ®£, “ - £ y o ^ S o c P^ ^ O _ >. s Cd o s ^ ^ S p c - 3 O 3 5 ^ 6 0 ( j b i O O c -c cd • c^ ^ — ■£ *g. • C o £ > g ^ J 3 >* o o c oft t- a w *E Si o" Si ccj b — c £ «— O 8 O 2 u « X_ — o /• a 0 cu 3 u ■-8 5 4>3 O o^— Cd S c - o = w •- c to > to 3 0) o c/? o 73 •— C c 0> o £ o ‘-5 n J!n 0 E J= ■§ «5 C 00 0 LU c «> 9 .s1u .E i g 3 ? ■B a 8 p u .a U 1 1 1 1 i-S 1 £ I I C B «g g pr3 y *o .= y 2 ir, a c. 5 2 J= 2 Ji E - •a .o' "O 1 i EL i 1 -O JS -I 3 5 u n p fc- 5 _ i S 2 7 s l I|.f5|s | 5 f . | I 'lf s ^ 1 1 ! = b o 1 g s - I % s §s M ft5 ooC55«25n-32 l l ' i I I I 5 0 O cC. rJ JS B - 5 js c £ .£ £l^CCjfcc _cd G *< S ^ O ^ v: c E °- c .“’ C Jft 2 SSiCr-fc-S c OOxcdP o . r a - a 9 - “ J c^t35 o5?^5wSc§0 ^ocO^' u^ftga-aoo^liS^ocO^' ™ § x §, 2 x 8 jz -a2-2«Ou?S: &1 I c i r u ,0 E-=i=~v5CraC= - - g g-^^p2.2- “ - 2 m 2 - a o £ a « a P « — 2 ,0 J2 w u u guSE“C^ u> a W s-W 2 “ 3 R-T3 C ^8 J o o S'0 o 1 « C s ^ 3 ^ C O u . ’ ■ “ .£ g .2 !! £ .g s j 5 5 rt a h a-c > c ■o5 y3 c m C p 60 O * o 25 • J as OJ as 5 c w .c ^ • = « 0 0 > , o ^ c .E - • O) .3 o ^ g r ^ r s « O.B c o ■a ■« Cl O fc- ft 2 , • E 2 o o w i % l ac u< n O OBn C9 '-3 —- p 2 >, ^ 0 a a c p U S T3 zB ^c c -a (Ncifti "5 a: 5 60 S E S w ^ ^ ^ 3s ’5L ^ I g I t-8 — — ? lo c > o 5 §5 ij ^! b y ae C l “53 o ^ _ ■§ t •S «£: -Si 2 i; c E w I I |S ■- E o p -g2 S -a cd s 5 - ~~o 0 o -o8 O o G. C 3 rj (N Si s O» O fcJi T3 2^ >O w). . g? ft -S c ri fj - ^ c .S '§ § '5 .2 .. «* O *— CUO ^ c ^ *3 +i 2 - j* c c £5.-0 "3 >^ »r> TJ- 'C-C <% -c ^ c o3 " 2 J*> ^ O G. § Cti£J w<•£ ?3 >> — *3*2 d c SQ 133 *2£ ' B PE ^«J2 ~5 -His *3e 'E.o 2 2 ■I s g. 2 £■ c ’■§ c -i I§ J •S g S ■3 u 2 2 "S. 'a S UJi/ H< - u *t; • - ’O O -53: Hr2 U 5 - .2— .=* c ,’ c £ v*i ^ ^5 3 a t < if z 5z 5 1 1 1 § I k > !s g 2 S-g E a- aisES •s* s S! * z ^ tJ5 o c 0- -'I *5 = t?« SPc ^5>% *53) u 1 £ 2 §.. e J§ <1 ^ 8 S< WUl/Bu i S * E % 0 & lie £s,a< s i t> E a !! E & C ■“ : 4 2 E.f < c: jr.E c 1 s 5.S«-B ■£ 3 c *2 2 3 c ^ S 182 3 c3 EE* •a §■ t 0-0 E-S € 7 U U 8 « C E £ ^ C 3 -Tl a ■5 8 a * JS 0.-5 S »G E c n *= “ cT §>> 1. “3 Msi 'O3 '5.&H 3 2 .1~ l ! f §1 > © n >> £ ■j II c L» -r t-n< —£ C ■ ||? d .2 *0 O S3 s. 1Z a 8 s| fl.i. a. SJj! *0 O 3 .ti, Z w 2 "g 1 3'§5 S « E K 3 >» ISE0 Ks «, £ E ■£ ■ o a C ■> ■*• 0 '£ -o c o -= . *3 T. z. ~- c I l l 3 w 2 u -- ! MS2 1 1-3 S’ tf .5 30 «J -a 2 £ •a o a* g| S 1 2 'c u •53 2c E5 .S O' « c c 9> C o* z i c. 0 • 3 3 a i i : UJ Z 05a £ 4S - 'I 7c a s0 c o.-g E £ T3 »' e §■§ S 5. JS -. 3 t«- E Co' -2 £ DCi t t S. - 'o -5J3 aj u-o tilUi §c I* 5 cu-S t i l S8S ■O U U E 2 .2 E ii-! Ji|l -C *3 2 6 | £ i n Z 0^ .1 I dc Sal z S *5.' u > 1 S 15 £Si'C OP I L— D | g Z — 1 s K 3 c & H i LU.5 of cffecl pacin : (rial 184-91. 2 z X C C , i “ |1 « ^ < 3 o y ^ S 'Ml zi |ll -§j D.' caHis? £ 2 23 .2 I I. „ £i f Wi¥ ^ ' Z “ 1 3 1 iia 3 X ' ._ ■£ O Is-; I CD 'o ^ K T* S -O S 1> K! «D E Sf- b ?j 5 g E oc 0 ^ 4 • -oc t■ l g-gl*S§ 1 § E 2 =* c c< £ j o ii .E a "2 3 ■j 3 = £*5 £I £ C ~ 2 'C ® §•'1 5 f g A | J 15 s.4 <«5. « V u '"si I u £ Pn 00-S e -25 « 8I E S . O £ ■ 2^t 7 ■ i £ V -■ £ -c~* | | ci ' 5 -S3 111 Si | £ 3: s I “ 2 - i> S 3 5,S| £ S •= § o = 5 ■ g ! _• -= OS2 C9 2 -> £5 11 “ c ^ I «§•£ i* I* I « Ig £ *.3 a* ‘5. £ «> -T.. S | g5 8 •5 2 K Is s I" §■ ■3 * > p S g -I a 1 Jog Is® ! a 3 £ tS-5 o* >> § ^ «, ’|«£ |-5 •S'§ uz -Sc £5 s- 4 .= .5 UO aO io X; te.2 | KC^ S-o? >- SMBE 5 1* I 5C 3 £ 4 *M £ i I I E f £ := -c 2 1! I . ~ 2 1U II Eij •-a" 'O E ~U 8 = S § 5 — C.2 CEl ^3 J ^ E# *5 c5 g'c -5^ c5f c*n 1 r* "" S > C/3 M . 2 ^ So 1 Uc wC ’3•« C 1 i “ IS s? z >• =;